The full syllabus
594 lessons in 17 tracks, from “what is AI” to training your own model. 8 are free on YouTube today; the rest say Coming soon until they go live.
Curriculum v1.0, 31 October 2026.
No lesson matches these filters. Clear filters or try the site search.
Track 0 AI Foundations (No Code)
What AI, ML, deep learning and GenAI are; how LLMs work; tokens, context, hallucination, safety and AI at work.
What AI IsPart of AI for Beginners (No Code): Understand and Use AI
- What Is AI, Really? Explain AI in one sentence and spot it in five apps you use daily
- AI vs ML vs Deep Learning vs GenAI, and Why AI Exploded Now Place any AI product on the AI > ML > DL > GenAI map and explain the boom
- How Machines Learn From Examples Describe learning from examples with a spam-filter story
- Data: The Fuel of AI Judge whether a dataset is good enough for an AI task
- Neural Networks Without Maths Explain layers, weights and training with one picture
- Training vs Using a Model Tell training from inference and why one costs crores and the other paise
- What AI Still Can't Do: Limits, AGI Claims and Hype Separate real capability from hype in a news headline
How LLMs WorkPart of AI for Beginners (No Code): Understand and Use AI
- What Is a Large Language Model? Explain an LLM as a next-word predictor trained on text
- Tokens and the Context Window Estimate tokens for a document and know when it won't fit
- How LLMs Write: Next-Token Prediction, Temperature, Sampling Predict how temperature changes an answer
- How ChatGPT Was Made: Pre-training, Fine-tuning, RLHF Explain the three training stages of a chat model
- Reasoning Models: What "Thinking" Really Means Choose when a reasoning model is worth the extra time and cost
- The Model Landscape: Closed vs Open Weights, Sizes, Benchmarks Read a model card and a leaderboard without being fooled
Trust, Hallucination and SafetyPart of AI for Beginners (No Code): Understand and Use AI
- Why AI Makes Things Up Explain hallucination and the three habits that catch it
- Using AI Safely Follow five safety rules for AI at home and at work
- Fact-Check and Bias-Check Any AI Answer Verify claims with sources and test for bias with paired prompts
- What Happens to Your Data in AI Chats Set the privacy and training settings in your AI apps
- Spot Deepfakes and AI Scams Use a 5-point check on suspicious voice, video and messages
- AI Rules You Should Know: EU AI Act, India DPDP, Copyright Know which rules touch your AI use at work
AI at WorkPart of AI for Beginners (No Code): Understand and Use AI
- AI Roadmap for Beginners Pick your next 30 days of AI learning
- 20 Everyday Work Tasks AI Can Do Today, and How to Measure the Gain Pick five tasks and measure time saved in a one-week test
- AI Agents vs Agentic AI: The Real Difference Explain agents vs agentic AI to a manager
- Workshop: AI Risk Review and Team AI Policy Run a risk review and ship a one-page AI policy for your team
Everyday AI Safety
- Voice-Clone Fraud: Family Code Words and Call-Backs Set a family code word and a call-back rule that defeat voice-clone scam calls
- Fact-Check Viral AI Images and Videos With Free Tools Trace a viral image or clip to its source with reverse search and content credentials
- What Not to Paste Into AI: A Privacy Checklist Redact personal, client and company data before you paste anything into a chatbot
- Copyright Basics: Who Owns AI-Made Work? Decide when you can use, sell or claim AI-made text, images and music
Track 1 Using AI Tools
ChatGPT, Claude, Gemini, Copilot, Perplexity, NotebookLM, image, video, voice and coding tools, with free vs paid comparisons.
Chat AssistantsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- How Chat Assistants Work Explain what happens between your message and the reply
- ChatGPT Complete Beginner Guide Use ChatGPT's free plan for writing, files and images
- How to Use Claude AI: Beginner's Guide (Free) Use Claude for long documents, writing and analysis
- Gemini Complete Beginner Guide Use Gemini with Gmail, Docs and YouTube
- Microsoft Copilot Guide: Free and Microsoft 365 Use Copilot in Windows, Edge and Office
- How to Use ChatGPT for Studying: 7 Prompts Study a chapter with 7 tested prompts
- ChatGPT vs Claude vs Gemini vs Copilot (+ Meta AI, Grok, DeepSeek): Same 5 Tasks Pick the best assistant per task from a real side-by-side test
- Free vs Paid AI Plans: What You Really Get Decide if a paid plan pays for itself
- Projects, Memory, Custom GPTs, Gems and Claude Projects Set up a reusable workspace for a recurring task
- Deep Research Modes Compared Produce a cited research brief and check every source
Research and Knowledge ToolsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- Perplexity Guide, and How It Compares to ChatGPT Search and Google AI Mode Get cited answers and pick the right AI search per question
- NotebookLM Guide: Study Systems, Work Knowledge Bases, Audio Overviews Turn PDFs and notes into a source-grounded knowledge base and podcast
- Cited Research Report Workflow Produce a 2-page report where every claim has a checked source
Office and ProductivityPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- AI in Excel and Google Sheets Clean, summarise and chart a sheet with AI
- AI for Email, Docs, Meetings and Slides Draft, summarise meetings and build a deck with built-in AI tools
- Analyse Real Spreadsheets With ChatGPT and Claude Get a correct analysis and verify every number
- AI Browsers and Agent Modes: What Works Today Hand a multi-step web task to an agent mode safely
Image ToolsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- How AI Makes Images, Audio and Video Explain how generators create media from text
- Free AI Image Generators Compared, Plus Editing (Is Midjourney Worth It?) Pick an image tool and edit results
- Image Prompting That Works Get the image you imagined in three tries or fewer
- Make a Consistent Brand Image Set Produce 10 on-brand images with a style guide
Video ToolsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- AI Video Generators and Avatars Explained Know what each video tool does and how to disclose avatars
- AI Video Editing: CapCut and Descript Cut a talking video by editing its transcript
- Build a Faceless Explainer Pipeline Produce a 3-minute explainer from script to upload
Voice and Audio ToolsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- AI Voice and Transcription Tools: Free vs Paid Turn a script into narration and transcribe audio for free
- AI Music: Suno and Udio, and the Licence Fine Print Make a background track and know if you can use it
- Turn Documents Into a Podcast Episode Produce an edited audio episode from source documents
Coding AssistantsPart of Browser, Computer-Use & Coding Agents
- AI Coding Tools Explained: Copilot, Cursor, Claude Code, Codex Pick a coding assistant for your level
- Build a Web App by Describing It: Lovable, Bolt, v0, Replit Ship a small working app without writing code
- Claude Code Free With Ollama: Full Setup Run Claude Code against a free local model
- Coding Assistants Compared on One Real Project Choose a coding agent from a measured head-to-head
Track 2 Prompt Engineering
From first prompts to tested, reusable prompt systems and context engineering.
Prompt BasicsPart of Prompt Engineering
- Prompting Basics Write a clear prompt that gets a usable first answer
- Better Prompts in 4 Simple Upgrades Upgrade any prompt with role, context, format and examples
- Few-Shot Prompting: Teach by Example Get consistent outputs with 2-5 examples
- Reasoning Prompts: Step-by-Step, Step-Back, Self-Check Cut errors on multi-step problems
- Get Tables and JSON Reliably Get output a spreadsheet or app can use
- Prompt Chaining: Break Big Tasks Into Steps Run a 4-step chain that beats one giant prompt
- System Prompts That Hold, and Resisting Prompt Injection Write a system prompt that survives tricky users and hidden instructions
Advanced Prompting and TestingPart of Prompt Engineering
- The One-Page Prompt Checklist Check any prompt against six points
- Reusable Prompt Templates and Variables Turn a good prompt into a template anyone can reuse
- Test Prompts Like Code Score two prompt versions on a 20-case test set
- Prompting Reasoning Models Is Different Adapt prompts for reasoning models
- Context Engineering: What Goes in the Window Design the full context for an assistant, not just the prompt
Prompts for WorkPart of Prompt Engineering
- Prompts for Emails, Reports and Summaries Write a weekly report in 10 minutes
- Extract Data From Messy Text Turn invoices or tickets into a clean table
- Build Your Team Prompt Library Ship a tested prompt library for one team
Track 3 Python for AI, From Zero
Setup, Colab and notebooks, Python basics, APIs, JSON, environment variables and keys.
SetupPart of Python for AI From Zero
- Why Python for AI, and What You Need Know the setup path for your machine
- Best Laptop for AI: Budget, Mid, Pro (India) Pick a laptop that runs local AI for your budget
- Install Python and VS Code (Windows, Mac, Linux) Run your first Python file
- Virtual Environments and Packages With uv Install packages without breaking other projects
- Git and GitHub for AI Projects Version and share a project on GitHub
- WSL and Docker Basics for AI Developers Run an AI tool in a container
Notebooks and Google ColabPart of Google Colab for AI: Free GPUs to Real Projects
- Colab Basics: Notebooks, Runtimes, Files, Sharing Create and share your first notebook
- Python and Data in Colab: Sheets, Drive, Charts Build a monthly report from a sheet
- Gemini Inside Colab: Generate, Explain, Fix Code Fix a broken notebook with AI and check the result
- Colab GPUs: T4, L4, A100 and What Costs Units Pick a GPU and measure the speed-up
- Run Open AI Models on Colab Without Breaking It Transcribe and summarise a meeting on a Colab GPU
- Colab Reliability: Version Clashes, Secrets, Disconnects Build a robust run-all template
- Colab Capstone: PDF or Recording to Summary and Actions Ship an end-to-end work tool notebook
Python FundamentalsPart of Python for AI From Zero
- Variables, Strings, Lists and Dictionaries Store and look up records in a script
- If, Loops and Conditions Process a list of items with rules
- Functions, Modules and Debugging Turn repeated code into functions and fix a traceback
- Files and Folders With pathlib Read and write text, CSV and JSON files
- Type Hints and Pydantic Models Validate data the way AI apps do
- Async Python for AI Apps Call five APIs in parallel
Data, JSON and APIsPart of Python for AI From Zero
- JSON for AI Developers Read and write the format every AI API uses
- Call an API From Python Fetch data from a public API
- Environment Variables and API Keys, Safely Keep keys out of code and out of GitHub
- pandas Basics for AI Work Clean and summarise a CSV
- Quick Interfaces: Streamlit and Gradio Put a web UI on a Python function
- Build an API With FastAPI Serve your code as a web API
- Testing With pytest Write tests that catch regressions
Data and SQL With AIPart of Python for AI From Zero
- Clean and Explore Messy Spreadsheets With AI Clean a messy sheet and get five checked insights from it with an AI assistant
- pandas With an AI Assistant: Ask, Check, Run Have AI write pandas code, then verify every number it reports
- Text-to-SQL: Ask Your Database in Plain English Turn plain-English questions into safe, read-only SQL and check the results
- Build a Chat-With-Your-Database Agent Build an agent that inspects tables, writes SQL, retries on errors and explains answers
- AI-Built Dashboards in Streamlit and Looker Studio Generate a working dashboard from a dataset and a one-paragraph brief
- Project: Sales Insights Assistant Ship an assistant that answers sales questions from a database with charts
Track 4 Working With LLM APIs
OpenAI, Anthropic, Gemini, open models, Ollama, llama.cpp, vLLM, Hugging Face; streaming, tool calling, structured output, cost and latency.
How AI Apps WorkPart of LLMs for Beginners (No Code): How AI Apps Really Work
- See How Every AI App Works Draw the request/response path of any AI app
- Know What Every AI Call Costs Estimate the cost of a feature before building it
- Control How the AI Answers Use system prompts and settings to shape answers
- Get Answers Your App Can Use Ask for structured output instead of prose
- Let AI Use Real Tools Explain tool calling end to end
- Stop AI Mistakes Reaching Users Add checks between the model and the user
ProvidersPart of LLM APIs Masterclass: OpenAI, Claude, Gemini & Local Models
- OpenAI API From Zero Make your first API call and read the response
- Anthropic Claude API From Zero Call Claude and handle messages and system prompts
- Gemini API From Zero (Free Tier) Call Gemini with a free key
- Enterprise Clouds: Azure OpenAI, Bedrock, Vertex Know what changes when your company uses a cloud provider
- Provider Shoot-Out on Your Own Task Pick a provider from measured quality, cost and latency
Open and Local ModelsPart of Open Models in Practice: Hugging Face, Ollama, llama.cpp & vLLM
- Open-Weight Models Explained: Llama, Qwen, Gemma, Mistral, DeepSeek Choose an open model and check its licence
- Run AI Locally With Ollama (and LM Studio) Chat with a local model and call it from Python
- Hugging Face for Beginners: Hub, Pipelines, Inference Providers Run a Hub model locally and via a hosted API
- llama.cpp, GGUF and How Big a Model You Can Run Run a quantized model on CPU and size models to your hardware
- Serve Models at Scale: vLLM (and SGLang, TGI) Serve an OpenAI-compatible endpoint on a GPU and pick an engine
Core API SkillsPart of LLM APIs Masterclass: OpenAI, Claude, Gemini & Local Models
- Messages, Roles and Multi-Turn Chat Keep a conversation going in code
- Streaming Responses and a ChatGPT-Style UI Ship a streaming chat app
- Tool Calling in Code Let the model call your Python functions
- Structured Output: JSON Schema, Pydantic, Invoice Extractor Get validated objects from every call
- Measure and Cut Cost and Latency Halve cost on a real feature with measurements
- Prompt Caching Cut repeat-context cost with provider caching
- Rate Limits, Fallbacks and Portable Code: LiteLLM, OpenRouter Keep an app running when a provider fails and swap models by config
LLM ProjectsPart of LLM APIs Masterclass: OpenAI, Claude, Gemini & Local Models
- Project: Document Summariser Summarise any PDF into key points
- Project: Cheap-First Model Router Route easy questions to a small model and hard ones to a big one
Ship an AI AppPart of LLM APIs Masterclass: OpenAI, Claude, Gemini & Local Models
- Next.js Front End for Your LLM Backend Connect a Next.js page to a Python LLM backend with typed requests
- Streaming Chat UI: Tokens, Stop and Retry Stream tokens into a chat UI with stop, retry and error states
- Add Login and Per-User Limits to Your AI App Add sign-in and a daily per-user request limit so nobody burns your API budget
- Deploy Your AI App on Free Tiers, No Card Needed Put the front end and backend online on free tiers with secrets kept safe
- Project: Ship a Public AI Study Helper Launch a logged-in, streaming AI study helper at a public URL
Track 5 RAG: Retrieval-Augmented Generation
Architecture, chunking, embeddings, every major vector database, retrieval types, every RAG type, frameworks, evaluation and production.
RAG ConceptsPart of RAG for Beginners (No Code): Make AI Answer From Your Files
- Why AI Can't See Your Files Explain the problem RAG solves
- RAG in One Picture Draw the ingest, retrieve, generate pipeline
- Chunks and Embeddings Explain chunks and embeddings without maths
- RAG Tutorial for Beginners: Free Python Chatbot Build a working RAG chatbot on a free local model
- Searching by Meaning Explain similarity search and top-k
- Writing the Answer Ground answers in retrieved text and cite sources
- RAG vs Fine-Tuning vs Long Context (and Cache-Augmented Generation) Choose the right approach for a use case
- Making RAG Good Diagnose the top five RAG failures
Ingestion and ChunkingPart of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- Loading PDFs, Docs and Web Pages Load a folder of mixed documents
- Chunking Basics: Fixed, Recursive and Structure-Aware Pick a chunking strategy with a quick test
- Hard PDFs: Tables, Scans and OCR Parse tables and scanned pages correctly
- Metadata, Filters and Self-Querying Filter retrieval by source, date and section
- Semantic and Advanced Chunking: Parent-Document, Contextual, Late Measure which strategy wins on your data
EmbeddingsPart of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- Embeddings Explained With Code, Free and Local Embed sentences on your laptop and compare them
- Choosing an Embedding Model: Paid vs Open, Multilingual, ColBERT Pick a model by quality, language and cost
- Fine-Tune an Embedding Model on Your Domain Lift retrieval accuracy with domain pairs
Vector Databases (One Lesson per Tool)Part of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- What a Vector Database Does: ANN, HNSW, IVF Explain indexes and the speed/recall trade-off
- Chroma Tutorial Store and query embeddings locally
- FAISS Tutorial Build an in-memory index for millions of vectors
- pgvector: Vectors in Postgres Add semantic search to an existing Postgres app
- Pinecone Tutorial Run a managed serverless index
- Qdrant Tutorial Use payload filters and hybrid search in Qdrant
- Weaviate Tutorial Use built-in vectorizers and hybrid search
- Milvus Tutorial Run a distributed vector DB for large collections
- Elasticsearch and OpenSearch for Vector and Hybrid Search Add vectors to an existing search stack
- Choose a Vector DB (incl. LanceDB, Redis, Atlas): Benchmark on Your Data Pick a DB from measured recall, latency and cost
Retrieval TypesPart of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- Dense Retrieval Retrieve by meaning with embeddings
- Sparse Retrieval and BM25 Catch exact terms, codes and names
- Hybrid Search Combine dense and BM25 with rank fusion
- Rerankers: Cross-Encoders, Cohere Rerank, BGE Re-score top results for precision
- Query Rewriting and Diversity: Multi-Query, HyDE, MMR Fix vague questions and stop duplicate chunks
RAG TypesPart of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- Naive RAG Build the baseline everything else improves on
- Advanced RAG: Rewriting, Reranking and Chat History Upgrade naive RAG and handle follow-up questions
- Modular RAG Assemble RAG from swappable modules
- Agentic RAG Let an agent decide what and when to retrieve
- Graph RAG: GraphRAG, LightRAG and Neo4j Answer whole-corpus questions with a knowledge graph
- Multimodal RAG: Images, Tables and Slides Retrieve from charts and images, not just text
- Corrective RAG (CRAG) Detect bad retrieval and fall back to web search
- Self-RAG Make the model critique its own retrieval and answer
- Adaptive RAG: Route by Question Type Route simple, complex and off-topic questions differently
RAG FrameworksPart of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- No-Code RAG: NotebookLM, Dify, Flowise Build a document assistant without code
- Chat With Your Documents in LangChain Build loaders, splitter, store and retriever in LangChain
- Chat With a PDF in 20 Lines (LlamaIndex) Build a query engine that shows sources
- Better Answers: Chunks, Metadata, Reranking (LlamaIndex) Measure three fixes side by side
- Many Documents, One Assistant (LlamaIndex) Route questions across multiple indexes
- Haystack Tutorial Build a RAG pipeline in Haystack
- LangChain vs LlamaIndex vs Haystack vs Plain Python Choose a framework for your team
RAG EvaluationPart of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- Why RAG Needs Evaluation Spot the difference between "looks right" and "is right"
- RAG Metrics: Faithfulness, Relevance, Recall@k Pick the metric for each failure
- RAGAS Tutorial, With a Synthetic Test Set Build a test set and score a RAG app
- Evaluate RAG With DeepEval Write RAG tests that run in CI
- Evaluate and Ship Your LlamaIndex Assistant Run faithfulness checks before release
- Component RAG Evals: Retriever, Reranker, Generator Find which RAG stage causes each wrong answer
RAG in ProductionPart of Advanced RAG: Vector Databases, Hybrid Search, GraphRAG & Agentic RAG
- From Notebook to Product: RAG Checklist List what changes between a demo and a product
- RAG Security: Permissions, PII, Poisoned Documents Enforce who can see which document
- Monitor and Trace RAG in Production (Phoenix, TruLens, Langfuse) Trace retrievals and alert on falling answer quality
- Scaling RAG: Fresh Indexes, Latency, Cost, Millions of Docs Plan incremental indexing, caching and sharding
- Capstone: Company Docs Support Bot Ship a cited, evaluated support bot over company docs
Track 6 AI Agents
From a simple agent to multi-agent architectures: the loop, tools, memory, planning, frameworks, MCP, human-in-the-loop, computer and coding agents, guardrails.
Agent FoundationsPart of AI Agents for Beginners (No Code): How Agents Really Work
- What Is an AI Agent? Tell an agent from a chatbot and a workflow
- How an AI Agent Keeps Going Explain think, act, observe, repeat
- What Can an AI Agent Touch? List an agent's tools and their risks
- Does Your AI Agent Remember You? Explain short- and long-term agent memory
- One AI Agent, or a Team? Decide single agent vs multi-agent
- Who Stops an AI Agent? Design stop rules and approvals
- Agent Design Patterns: Chaining, Routing, Parallel, Orchestrator, Evaluator Map a business problem to a proven pattern
- Tool Calling Explained (No Code) Explain how a model asks for a tool, your app runs it and the result comes back
Build an Agent From ScratchPart of Build Your Own Agent Harness
- What an Agent Harness Really Is Draw the loop real coding agents use
- The Agent Loop in 50 Lines Write a working agent loop with one tool on a local model
- Tools and Permissions Add input checks, ask-before-acting and a sandbox folder
- Context: What the Agent Remembers Budget tokens with summaries and notes files
- Plans, To-Do Lists and Sub-Agents Make the agent plan and delegate to a helper
- Safety, Evals, and Rebuilding With an SDK Guard, test and port your agent to an official SDK
Tools, Reasoning and PlanningPart of AI Agent Frameworks: CrewAI, OpenAI Agents SDK, Claude Agent SDK, ADK, PydanticAI
- Design Tools Agents Use Correctly Write tool names, descriptions and schemas that work
- ReAct: Reason + Act Implement ReAct and read its trace
- Plan-and-Execute and Long-Running Agents Separate planning from doing and keep long tasks on track
- Reflection and Self-Critique Add a critic step that improves answers
Agent MemoryPart of AI Agent Frameworks: CrewAI, OpenAI Agents SDK, Claude Agent SDK, ADK, PydanticAI
- Agent Memory Types: Short, Long, Episodic, Semantic, Procedural Pick the memory type for a use case
- Long-Term and Graph Memory: Vector Stores, Mem0, Letta, Zep Recall user facts and relationships across sessions
- Conversation Memory, Compaction and Forgetting Keep long chats in budget and decide what to forget
Agent Frameworks (One Lesson Each)Part of AI Agent Frameworks: CrewAI, OpenAI Agents SDK, Claude Agent SDK, ADK, PydanticAI
- Give It Tools: Your First LangChain Agent Build a tool-using agent with LangChain
- OpenAI Agents SDK Build an agent with handoffs and guardrails
- Claude Agent SDK Build an agent on the harness behind Claude Code
- CrewAI Build a role-based crew of agents
- AutoGen / AG2 (and Microsoft's Agent Stack) Build a conversational multi-agent system
- smolagents Build a code-acting agent in a few lines
- PydanticAI Build a type-safe agent with validated outputs
- Google Agent Development Kit (ADK) Build and run an agent with ADK
- Agents and Workflows in LlamaIndex Give your data an agent
- Same Agent, Four Frameworks (+ TypeScript: Mastra, Vercel AI SDK) Compare code size, control, cost and traces
LangGraph Deep DivePart of LangGraph Builder
- Why Graphs? State, Nodes and Edges Explain chain vs graph and when to use which
- Your First Graph in 15 Minutes Run a three-node workflow on a local model
- Loops and Decisions Build a writer that checks and fixes its own draft
- Build an Agent Loop by Hand Implement tool calls inside a graph
- Memory and Human Approval: Interrupts and Resume Add checkpoints and approve-before-acting
- Multi-Agent Teams and Debugging Build a supervisor with two workers and debug it
Model Context Protocol (MCP)Part of Build Your First MCP Server
- MCP Explained Explain hosts, clients and servers in plain words
- How MCP Really Works, Before You Code Read a real MCP message exchange
- Your First MCP Server in 10 Minutes Build and test a two-tool server
- Give Your AI a Notes App (MCP) Build a notes server with good tool schemas
- Beyond Tools: MCP Resources and Prompts Choose tool vs resource vs prompt
- Connect Your MCP Server to a Real API Handle errors, timeouts and secrets
- Agent-to-Agent Protocols (A2A) Let agents from different vendors talk
Multi-Agent PatternsPart of AI Agent Frameworks: CrewAI, OpenAI Agents SDK, Claude Agent SDK, ADK, PydanticAI
- Why Multi-Agent, and When Not To Know the cost and benefit of more agents
- Supervisor and Orchestrator-Workers Patterns Route and split work across specialist agents
- Swarm and Handoffs Let agents hand off to each other
- Hierarchical Teams, and Why Multi-Agent Systems Fail Build teams of teams and diagnose loops and drift
- Debate and Critique Agents Improve answers with adversarial agents
- Project: Deep Research Agent Team Ship a team that searches, drafts, cites and fact-checks
Human-in-the-Loop and GuardrailsPart of AI Security: Prompt Injection, OWASP LLM Top 10 & Guardrails
- Human-in-the-Loop Explained Decide which actions need approval
- Guardrails: Inputs, Outputs, Sandboxes, Least Privilege Block unsafe inputs/outputs and sandbox agent actions
- Evaluate Agents: Trajectories and Tool Use Score an agent's steps, not just its answer
Computer, Browser and Coding AgentsPart of Browser, Computer-Use & Coding Agents
- Computer-Use and Browser Agents Explained Know what screen-driving agents can and can't do
- Coding Agents Explained: Claude Code, Codex, Copilot Agent Know how coding agents plan, edit and test
- Build a Browser Agent With browser-use and Playwright Automate a multi-page web task
- Working With Coding Agents: Rules, Plans, Tests, CI Get reliable changes locally and in GitHub Actions
- Computer-Use APIs Drive a desktop app with a computer-use model, safely
Agent Harness EngineeringPart of Build Your Own Agent Harness
- The Eight Parts of an Agent Harness Name the loop, tools, context, memory, permissions, sandbox, subagents and hooks in any agent product
- How Coding-Agent Harnesses Are Built (Claude Code, Codex) Compare coding-agent harness designs from their public docs
- Skills and Plugins: Package Know-How for Agents Write a skill an agent loads only when it needs it
- Hooks: Run Your Own Code Before and After Every Tool Call Block, log or fix tool calls with hooks
- Human Approvals: Permission Modes and Allow/Deny Rules Design an approval policy that is safe without nagging
- Context Engineering for Long Tasks: Compaction and Notes Keep a multi-hour task on track across context resets
- Sandboxing Agents: Containers, Files, Network, Secrets Run an agent with least privilege in a container
- Project: Build a Mini Coding-Agent Harness Ship a small harness that fixes a failing test in a sandbox
- Evaluate a Harness: Same Model, Different Harness Measure task success, pass^k and cost per solved task across harness changes
- Cost and Latency Budgets for Agents Set and enforce per-task cost and time budgets
Track 7 Automation and No-Code Agents
n8n, Make, Zapier AI, Power Automate, Google Apps Script plus AI, no-code agent builders and work automations.
Automation BasicsPart of n8n AI Agents & Automation for Work
- Automation 101: Triggers, Actions, and Where AI Fits Map one weekly task into trigger and actions
- Zapier AI Tutorial Build an AI zap that summarises new emails
- Make Tutorial With AI Build a scenario with an AI step
- n8n for Beginners Build your first n8n workflow on the cloud or locally
- Power Automate With AI Automate a Microsoft 365 task with AI Builder/Copilot
- Google Apps Script + Gemini API Add AI to Sheets and Gmail with a script
- n8n vs Make vs Zapier vs Power Automate Pick a platform by cost, limits and hosting
- Self-Host and Harden n8n: Docker, Errors, Retries, Monitoring Run n8n on your server and keep workflows running unattended
n8n AI AgentsPart of n8n AI Agents & Automation for Work
- Your First AI Agent in n8n Build an agent node with two tools
- RAG in n8n: Chat With Your Drive Files Build a document chatbot without code
- Project: AI Email Triage Classify, draft and route incoming email
- n8n and MCP Expose workflows as tools for any AI assistant
- Multi-Agent Workflows in n8n Chain specialist agents in one workflow
No-Code Agent BuildersPart of n8n AI Agents & Automation for Work
- Custom GPTs, Gems and Claude Projects as Simple Agents Build a no-code assistant for one job
- Microsoft Copilot Studio Build a company assistant in Copilot Studio
- Dify, Flowise and Langflow Build an agent app visually
- When No-Code Hits Its Limits Know when to move to code
Work Automation ProjectsPart of n8n AI Agents & Automation for Work
- Project: Automated Weekly Report Send a weekly AI-written report from a sheet
- Project: Invoice Processing and Lead Enrichment Extract invoices to a sheet and enrich new leads
- Project: Customer Support Agent With Handoff Answer FAQs and hand off to a human
Track 8 Voice AI Agents
Speech-to-text, text-to-speech, real-time voice agents, telephony, and training your own voice.
Voice FoundationsPart of Production Voice AI Agents
- How Voice Agents Work: STT, LLM, TTS Draw the voice agent pipeline
- Speech-to-Text: Whisper, Deepgram, AssemblyAI Pick an STT engine by accuracy, cost and language
- Text-to-Speech for Agents Pick a TTS voice by latency and quality
- Latency, Turn-Taking and Interruptions Get a voice agent under one second to first audio
- Speech-to-Speech Realtime APIs Build a voice agent on a realtime model
- Voice AI for Indian Languages Test Hindi and regional STT/TTS
- Voice UX: Prompts, Personas, Fallbacks Design a voice flow callers finish
Build Voice AgentsPart of Production Voice AI Agents
- Your First Voice Agent, No Code (and Hosted Platforms Compared) Launch a voice agent and decide build vs buy
- Build a Voice Agent With Pipecat Run an open-source voice pipeline
- Build a Voice Agent With LiveKit Agents Run a real-time voice agent in the browser
- Connect a Voice Agent to a Phone Number Take real calls via a telephony provider
- Voice Agents With Tools and RAG Book appointments and answer from docs on a call
- Test, Monitor and Deploy Voice Agents Run a voice agent reliably with simulated-call tests
- Capstone: AI Receptionist Ship a receptionist that answers, books and hands off
Train Your Own AI VoicePart of Train Your Own AI Voice for Free (Open-Source, Fine-Tuned)
- How Modern TTS Works, and Licences Check if a TTS model allows your use
- Zero-Shot Voice Cloning in 10 Minutes Clone your voice from a 20-second sample
- Responsible Voice Cloning: Consent, Watermarks, Disclosure Fill a disclosure checklist
- Build a Voice Dataset Record, cut and transcribe a clean dataset
- Make It Sound Pro Polish narration with takes, pauses and pronunciation fixes
- Fine-Tune a TTS Model on Colab Train and pick the best checkpoint before overfitting
- Production Narration Pipeline Voice a 10-minute script in one batch
- Capstone: Your Voice, End to End Narrate a 2-minute explainer and blind-test it
Track 9 LLM Engineering and Production
Testing and evals, observability, cost control, security, deployment, caching and routing.
Production FoundationsPart of Production AI for Beginners (No Code): Run AI for Real Users
- From Demo to Product List what breaks when real users arrive
- Cost and Speed Estimate monthly cost and latency of an AI feature
- Your AI Roadmap Plan your path from user to builder
- Measuring Quality Define what "good" means for an AI feature
- Safety and Privacy Spot privacy and safety risks before launch
- Running Reliably Plan fallbacks, retries and monitoring
- LLM System Design Design an AI system end to end on a whiteboard
Testing and EvalsPart of Agent Testing & Evaluation v2 (update of the live DeepEval course)
- Why Evals Are the Unit Tests of AI Write your first five eval cases
- Build an Eval Dataset (and Compare Prompts With promptfoo) Collect cases and run config-driven comparisons
- LLM-as-Judge, Done Right Write a judge prompt and check it against humans
- DeepEval Tutorial Run LLM tests with pytest
- Regression Gates in CI Block a release when quality drops
- Evaluating Agents End to End Score tool use, trajectories and outcomes
ObservabilityPart of Agent Observability & Cost Control
- Why AI Apps Need Observability Know what to log for every call
- Tracing LLM Apps With OpenTelemetry Trace a request across model and tool calls
- Langfuse and LangSmith Tutorial Trace, score and debug an app in Langfuse or LangSmith
- Dashboards and Alerts for AI Apps Alert on cost, latency and quality drift
Cost, Caching and RoutingPart of Agent Observability & Cost Control
- Where the Money Goes: Token Accounting Attribute cost per user and per feature
- Caching: Prompt Caching and Semantic Caching Serve repeat questions from cache
- Model Routing, Cascades and Latency Engineering Send each request to the cheapest model that passes, faster
- AI Gateways, Budgets and Kill Switches Centralise keys, logs, limits and spend caps
SecurityPart of AI Security: Prompt Injection, OWASP LLM Top 10 & Guardrails
- AI Security Basics and Data Leaks (PII, Secrets, Logs) List the main attacks and redact PII before the model
- Prompt Injection: Direct and Indirect Reproduce and mitigate an injection attack
- OWASP Top 10 for LLM Applications Check an app against the current OWASP list
- Agent and MCP Security: Excessive Agency, Tool Poisoning Lock down an agent's tools and servers
- Guardrail Tools: Llama Guard, NeMo Guardrails, Guardrails AI Add a safety classifier to an app
- Red-Team Your AI App: garak and PyRIT Run an automated attack scan and fix findings
DeploymentPart of LLM APIs Masterclass: OpenAI, Claude, Gemini & Local Models
- Wrap Your AI App in an API Serve an LLM feature over HTTP
- Dockerise an AI App Build and run your app in a container
- Deploy: HF Spaces, Render, Cloud Run, Lambda, Modal Put an app online on a free tier and scale to zero
- CI/CD for AI Apps Test, evaluate and deploy on every push
- Self-Host an Open Model on a GPU Cloud (and When Kubernetes Helps) Serve your own model with vLLM
- Capstone: Ship a Production AI Service Deploy an evaluated, traced, secured AI service
Evals in DepthPart of Agent Testing & Evaluation v2 (update of the live DeepEval course)
- Eval-Driven Development: Write the Test Before the Prompt Turn a feature idea into pass/fail checks before changing a prompt
- Error Analysis First: Read the Traces, Then Pick Metrics Name your app's top failure modes from 50 real outputs
- Golden Datasets and Test Sets That Last Build a versioned golden set with dev/test splits and edge cases
- Metrics for LLM Apps: From Code Checks to Rubrics Choose the cheapest metric that catches each failure mode
- Tool-Call and MCP Contract Tests Assert the right tool, the right arguments and safe handling of tool errors
- Safety Evals: Jailbreaks, Refusals and Over-Refusal Measure unsafe-answer and over-refusal rates on a fixed test set
- Human Review Loops: Annotation Queues and Rubrics Run a weekly review queue that grows the golden set
- Calibrate Your LLM Judge: Human Agreement and Bias Measure judge-human agreement and fix the biases that skew scores
- Agent Task-Success Evals: End-State Checks and pass^k Score agents on finished tasks, not nice-sounding answers
- Test Multi-Agent Systems: Handoffs, Loops and Who Broke It Catch bad handoffs and runaway loops between agents
- Flaky Evals in CI: Repeats, Thresholds, Baselines Gate releases on real regressions, not random noise
- Online Evals and A/B Tests on Live Traffic Score sampled production traces and run a safe A/B rollout
- Benchmarks and Their Limits: Contamination, Saturation Read a leaderboard critically and build a private benchmark instead
- Capstone: The Agent Quality Platform Ship datasets, metrics, red-team tests, CI gates and live monitoring for one agent
Eval Tools (One Lesson Each)Part of Agent Testing & Evaluation v2 (update of the live DeepEval course)
- promptfoo: Your First Eval Suite in YAML Compare prompts and models side by side from one config file
- Inspect (UK AISI): Tasks, Solvers and Scorers Write an Inspect task and read its log viewer
- OpenAI Evals: Open-Source Framework and Paid Evals API Run a graded eval on the framework and on the API (verify price)
- LangSmith Evaluations: Datasets and Experiments Run and compare experiments in LangSmith (paid plans; free tier limits: verify price)
- Langfuse Evals: Datasets, Experiments, LLM Judges Score traces and dataset runs in self-hosted Langfuse
- Braintrust: Experiments, Scorers and autoevals Run experiments with autoevals scorers (paid platform; free tier: verify price)
- Arize Phoenix Evals: Trace First, Then Evaluate Evaluate spans from a traced agent locally
- TruLens: Feedback Functions and the RAG Triad Score context relevance, groundedness and answer relevance
- DeepEval Advanced: Custom Metrics and Agent Tracing Write custom metrics and trace-level agent evals in DeepEval
- RAGAS Advanced: Agent, Tool-Call and Multi-Turn Metrics Evaluate agent workflows and conversations with RAGAS
- One Eval, Six Tools: Pick Your Eval Stack (Free vs Paid) Choose a stack by cost, hosting, CI fit and team size
Track 10 Training LLMs
How training works, data, fine-tuning (full, LoRA/QLoRA), instruction and preference tuning, distillation, quantization, evaluation, tools, small models from scratch, and speech-model fine-tuning.
How Training WorksPart of Fine-Tune Your Own LLM (Free GPU, Step by Step)
- Fine-Tune vs Prompt vs RAG Decide for three real work scenarios
- How LLM Training Works: Tokens, Loss, Pre-training vs Fine-tuning Read a loss curve and explain it
- GPUs for Training: Colab, Kaggle, Cloud GPUs Safely Pick hardware and shut it down after the job
- Pre-training at Scale, Explained Explain data mixtures, tokens seen and compute budgets
- Scaling Laws and Compute-Optimal Training Estimate model size vs data for a budget
Training DataPart of Fine-Tune Your Own LLM (Free GPU, Step by Step)
- Chat and Instruction Data Formats Format examples the trainer expects
- Build a 300-1,000 Example Dataset: Clean, Split, Licence-Check Collect, clean and split a dataset without leakage
- Synthetic Data With Guardrails Generate and filter synthetic examples
- Deduplication and Decontamination Keep test data out of training data
Fine-Tuning MethodsPart of Fine-Tune Your Own LLM (Free GPU, Step by Step)
- Full Fine-Tuning vs Parameter-Efficient (PEFT) Know why LoRA fits on one GPU
- LoRA Explained and Applied Train a LoRA adapter
- QLoRA: Bigger Models on Small GPUs Fine-tune a 4-bit model on a free GPU
- Your First Fine-Tune on Colab (Instruction Tuning) Instruction-tune a model and resume after disconnect
- Make It Better: Data, Hyperparameters, Continued Pre-training, Multi-GPU Run a second training that beats the first
Preference Tuning and AlignmentPart of LLM Training Beyond Basics: Preference Tuning, Evals and Deployment
- RLHF and Preference Tuning, Explained Explain why chat models feel helpful
- DPO Hands-On (and ORPO, KTO, Reward Models) Align a model with preference pairs and pick a method
- GRPO and RL for Reasoning Train a small model to reason with verifiable rewards
Training Tools (One Lesson Each)Part of Fine-Tune Your Own LLM (Free GPU, Step by Step)
- Hugging Face Transformers for Training Load, tokenize and train with Trainer
- Hugging Face TRL and PEFT Run SFT and DPO with LoRA adapters
- Unsloth (and LLaMA-Factory) Fine-tune faster on one GPU
- Axolotl Run config-driven fine-tunes
- torchtune Fine-tune with PyTorch-native recipes
Evaluate, Compress and ShipPart of Fine-Tune Your Own LLM (Free GPU, Step by Step)
- Benchmarks Explained, and lm-evaluation-harness Read benchmark tables critically and run standard benchmarks
- Evaluate Your Fine-Tune Properly Produce a before/after table on a held-out set
- Quantization: GGUF, GPTQ, AWQ, bitsandbytes Shrink a model and measure the quality hit
- Merge, Export, Run Your Model (and Model Merging) Run your fine-tune in Ollama behind an API
- Distillation: Teach a Small Model From a Big One Distil a task from a large model to a small one
- Capstone: Fine-Tune for a Real Work Task Deliver a model, eval report and demo
- Model Merging With mergekit: SLERP, TIES, DARE Merge two fine-tunes into one model on a free GPU and test that it improved
Small Models From ScratchPart of Transformers & GPT From Scratch for AI Engineers
- How a GPT Is Built, End to End Explain every stage from text to chatbot
- Build a BPE Tokenizer Train a tokenizer on your own text
- Build and Train a Toy GPT Train a tiny GPT that writes in a style
- Train a Small Language Model on Colab Pre-train and evaluate a small LM on a curated corpus
Speech Model Fine-TuningPart of Train Your Own AI Voice for Free (Open-Source, Fine-Tuned)
- How Speech Models Are Trained Explain ASR and TTS training in plain words
- Fine-Tune Whisper for Indian Accents Cut word error rate on your speakers
Track 11 ML and Deep Learning Foundations for AI Engineers
Just enough maths intuition, classic ML, neural networks, transformers, attention and embeddings.
Maths IntuitionPart of Transformers & GPT From Scratch for AI Engineers
- Vectors, Similarity and Probability, Visually Compute cosine similarity and read a token distribution
- Matrices: What a Layer Really Does Read a matrix multiply as "mixing features"
- Gradients, Loss Functions and Gradient Descent Explain how a model learns by going downhill
- Optimisers: SGD, Adam, Learning-Rate Schedules Tune a learning rate from a loss curve
Classic MLPart of Transformers & GPT From Scratch for AI Engineers
- Supervised vs Unsupervised Learning Frame a business problem as an ML task
- Your First Model With scikit-learn Train and evaluate a classifier
- Metrics and Overfitting: Precision, Recall, F1, Validation Pick a metric and spot overfitting
- Trees and Gradient Boosting Beat a baseline on tabular data
Neural NetworksPart of Transformers & GPT From Scratch for AI Engineers
- From One Neuron to a Network Build a neuron in NumPy
- Backpropagation, Step by Step Trace gradients through a tiny network
- PyTorch Basics Train a model with a PyTorch training loop
- Project: Digit Recogniser Train and test a network on MNIST
- Before and Around Transformers: CNNs, RNNs, Regularisation Know why transformers replaced RNNs and stabilise training
Transformers and AttentionPart of Transformers & GPT From Scratch for AI Engineers
- Embeddings: Meaning as Numbers Explain why king - man + woman is close to queen
- The Transformer in One Picture Name every block of a transformer
- Self-Attention, Multi-Head and Positions, Step by Step Compute attention for a 3-word sentence
- Code a Transformer Block Implement a decoder block in PyTorch
- KV Cache and Why Long Context Is Expensive Estimate memory for a context length
- Mixture of Experts, Beyond Transformers, and Looking Inside Models Explain MoE, the alternatives and interpretability basics
- Interpretability Basics: Look Inside a Model Inspect attention and activations in a small model to see what drives an answer
Track 12 Multimodal AI
Vision, image generation, video and speech, for builders.
VisionPart of Browser, Computer-Use & Coding Agents
- How AI Sees: Vision Models Explained Explain how a model reads an image
- Image Understanding With GPT, Claude and Gemini APIs Extract data from photos and screenshots
- Document AI: OCR, Forms and Tables Turn scanned forms into JSON
- Object Detection With YOLO Detect and count objects in images
- Open Vision-Language Models Locally Run a VLM on your own GPU
- Project: Receipt and ID Checker Build a validated image-to-data pipeline
Image Generation for Builders
- How Diffusion Models Make Images Explain noise-to-image in plain words
- Generate Images From Code via APIs Generate and edit images in a script
- Stable Diffusion, FLUX and ComfyUI Locally (+ ControlNet) Run and control open image models
- Train an Image LoRA on Your Style or Product Fine-tune an image model on 20 images
Video AI for Builders
- How AI Video Generation Works Explain video diffusion and its limits
- Generate Video From Code: APIs and Open Models Generate a clip via API or an open model
- Project: Automated Explainer Video Pipeline Script, voice, visuals and edit in one pipeline
Speech and Audio for BuildersPart of Production Voice AI Agents
- Speech AI for Builders Map STT, TTS, diarization and audio generation
- Whisper Locally With Speaker Diarization Transcribe hours of audio and label speakers
- Project: Meeting Assistant Transcribe, summarise and assign action items
Track 13 AI Careers and Portfolio
Roles, learning plans, portfolio projects, interviews, freelancing, and AI for your current job.
AI Roles
- AI Jobs Explained: AI Engineer vs ML Engineer vs Data Scientist Pick a target role
- Forward Deployed Engineer: The New AI Job Know what FDEs do and how to become one
- AI Engineer Roadmap Plan 6 months of AI engineering learning
- AI Product Manager: Skills and Path Write an AI feature spec with evals
Learning and Credentials
- Best AI Certifications: Worth It or Skip? Pick a certification that helps your goal
- How to Read an AI Paper Summarise a paper's claim, method and evidence
Portfolio
- What Makes an AI Portfolio Hireable (GitHub, Demos, Open Source) Plan three projects and present them well
- 20 AI Portfolio Project Ideas by Role Pick projects matched to a job
- Capstone: Three-Project Portfolio Ship three evaluated, deployed projects
Interviews
- AI Job Interviews: What to Expect Know the rounds and how to prepare
- 50 LLM, RAG and Agent Interview Questions Answer core concept questions clearly
- AI System Design Interviews and Take-Homes Design a RAG or agent system live and deliver a take-home
Freelancing and Consulting
- Freelancing With AI Skills: Where to Start Pick a service you can sell this month
- Packaging and Pricing AI Services Write a one-page offer with a price
- Finding Clients, and Growing Into an AI Agency Run a 30-day outreach plan
AI for Your Current JobPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- AI for Finance, Accounting and Analysts Automate reconciliations and turn data into decisions safely
- AI for HR Draft JDs, screen fairly and build onboarding
- AI for Marketing and Sales Plan campaigns and personalise outreach
- AI for Students and Teachers, Honestly Study and teach faster without cheating
- AI for Software Developers Use coding agents in a real team workflow
- AI for Chartered Accountants and Tax Professionals (India) Speed up compliance work and verify outputs
- Project: Automate 5 Hours a Week of Your Job Ship one automation and measure time saved
Track 14 Tools Directory: Free vs Paid vs Open-Source
Every category of AI tool with free, paid and open-source options; India pricing notes marked 'verify price' unless checked.
Everyday ToolsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- Free vs Paid vs Open-Source: How to Choose Use a 5-question checklist before paying
- Chat and Research Assistants: Free vs Paid Pick an assistant plan (verify price)
- Free AI for Students Claim legitimate student offers
- Image, Video and Voice Tools Directory Pick media tools by budget and licence (verify price)
- Paying for AI From India: INR Plans, GST, UPI, Cards Know the real INR cost before subscribing (verify price)
Builder Tools
- Coding Assistants Directory Pick a coding assistant tier (verify price)
- Automation Platforms Directory Pick by free-tier limits and self-hosting
- LLM APIs and Local Runtimes Directory Compare APIs, free tiers and local runtimes (verify price)
- Vector Databases Directory Pick a DB by scale, hosting and cost
- Agent Frameworks Directory Pick a framework by language, control and ecosystem
Infrastructure Tools
- Free GPUs, Notebooks and GPU Clouds (India Notes) Pick Colab, Kaggle or a cloud GPU (verify price)
- Eval, Observability and Security Tools Directory Pick eval, tracing and guardrail tools
- Deployment Platforms Directory Pick a host by free tier and scale (verify price)
- Training Tools Directory Pick a fine-tuning stack
Track 15 Frontier and Advanced
For builders who have shipped an agent: memory systems, agent red teaming and MCP security, the 2026 MCP spec and A2A, reasoning models and inference optimisation, reliability, RL for agents, knowledge graphs and AI governance (India DPDP basics).
Agent Memory SystemsPart of AI Agent Frameworks: CrewAI, OpenAI Agents SDK, Claude Agent SDK, ADK, PydanticAI
- Why Agents Forget: Memory as a System, Not a Prompt Explain write, store, retrieve and forget for agent memory
- Build Episodic and Semantic Memory From Scratch Store events and facts, retrieve them by relevance and recency
- Mem0 Hands-On: Extract, Update and Scope Memories per User Add per-user long-term memory to an agent
- Letta (MemGPT-Style) Agents: Self-Editing Memory Build an agent that manages its own memory
- Test and Govern Agent Memory: Recall, Staleness, Deletion Prove memory helps, and delete it on request
Agent Security and Red TeamingPart of AI Security: Prompt Injection, OWASP LLM Top 10 & Guardrails
- How Agents Get Attacked: The Risky Combination Spot the risky combination in any agent design
- Agent Defences That Hold Against Prompt Injection Apply design patterns that limit what an injected prompt can do
- MCP Security Lab: Tool Poisoning, Tokens, OAuth Scopes Attack and then harden a test MCP setup
- Red-Team Agents With PyRIT, garak and promptfoo Run an automated red-team campaign against a tool-using agent
MCP and A2A, AdvancedPart of Build Your First MCP Server
- Local vs Remote MCP Servers, Explained Choose stdio or a remote server and know what changes
- MCP 2026-07-28 Spec: Stateless, Discover, Apps, Tasks Explain what the new spec changes and migrate a server
- Remote MCP Servers With OAuth, Step by Step Protect a remote MCP server with OAuth
- Build With MCP Apps and Tasks Ship a tool with a UI and a long-running task
- A2A Advanced: Agent Cards, Tasks, Streaming, Auth Expose an agent over A2A with auth and streaming
- Project: Two Agents, Two Frameworks, One A2A Link Connect agents built in two frameworks through A2A
Reasoning Models and InferencePart of Open Models in Practice: Hugging Face, Ollama, llama.cpp & vLLM
- Reasoning Models in Production: When They Pay Off Decide per task whether a reasoning model is worth it
- Thinking Budgets: Cost, Latency and Quality Tune reasoning effort against an eval set (verify price)
- Long Context vs RAG: Measure Cost and Accuracy Measure when to stuff the context and when to retrieve
- Small and On-Device Models: Laptops, Phones and Browsers Pick and run a small model that fits your device
- Inference Optimisation: KV Cache and Prefix Caching Explain and measure what each optimisation saves
- Speculative Decoding and Continuous Batching Turn on speculative decoding and batching and measure the speed-up
- Project: Benchmark Your Model Serving Stack Compare two serving setups on your own GPU budget (verify price)
Frontier Agent EngineeringPart of AI Agent Frameworks: CrewAI, OpenAI Agents SDK, Claude Agent SDK, ADK, PydanticAI
- Why Agents Fail in Production, and Service Levels List the failure modes and set targets for an agent
- Agent Reliability: Retries, Idempotency, Checkpoints Make tool calls safe to retry and agent runs resumable
- Multimodal Agents: Screens, Documents and Voice in One Loop Build an agent that reads images and documents and acts on them
- Knowledge Graphs for Agents: Query and Update Tools Give an agent a graph it can query and update
- Durable Long-Running Agents: Resume After a Crash Run an agent for hours and survive restarts
- RL Fine-Tuning for Agents: GRPO on Tool Use Train a small model to call tools better with GRPO
- Project: Agent With Knowledge-Graph Memory Ship an agent that builds and queries its own graph
AI Governance at WorkPart of AI Security: Prompt Injection, OWASP LLM Top 10 & Guardrails
- Write an AI Use Policy for Your Team Draft a one-page policy: approved tools, data rules, review steps
- India DPDP Act Basics for AI Teams Map an AI feature to DPDP Act duties and the phased DPDP Rules 2025 timeline
- Privacy Engineering for AI Apps: PII and Retention Redact PII and set retention before data reaches a model
- AI Risk Management: NIST AI RMF and ISO/IEC 42001 Run a lightweight risk review for a new AI feature
- Audit Trails for AI Agents Keep the records you need when an agent gets something wrong
Track 16 AI for Your Role
Practical AI paths for students, teachers, small business owners, marketers, finance and HR/operations teams (free tools first), plus an advanced strategy module for leaders.
AI for StudentsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- Study With AI Without Cheating Yourself Set up AI as a tutor that quizzes and explains instead of writing for you
- Turn Lecture Notes Into Flashcards and Quizzes Convert a week of notes into flashcards, a practice quiz and a summary
- Research Assignments With AI and Real Citations Find, read and cite real sources, and catch invented references
- Project: Build Your Exam Revision Kit Build a personal revision kit for one subject: notes, quizzes and a study plan
AI for TeachersPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- Lesson Plans and Worksheets in Minutes With AI Draft a lesson plan, worksheet and answer key for any topic and grade
- Faster, Fairer Feedback on Student Work Use a rubric prompt to draft feedback you check and personalise
- Set Classroom AI Rules Students Will Follow Write clear AI-use rules and assignments that are hard to outsource
AI for Small Business OwnersPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- AI for Small Business: Your First Week Pick three daily tasks to hand to free AI tools this week
- Answer Customer Messages Faster With AI Build a reply library and tone guide for WhatsApp, email and reviews
- Product Photos, Descriptions and Posts With AI Create catalogue photos, descriptions and social posts for ten products
- Project: Automate Orders and Follow-Ups Log orders to a sheet and send follow-up messages automatically
AI for Marketing and ContentPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- A Month of Content From One Brief Turn one campaign brief into a 30-day content calendar with drafts
- Keep Your Brand Voice With a Custom Assistant Build a custom assistant that writes in your brand voice every time
- Keyword and Audience Research With AI Map audience questions and search keywords, then check them against real data
- Project: Repurpose One Video Into 10 Posts Turn one long video into clips, captions, a blog post and a newsletter
AI for Finance and AccountingPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- Reconcile and Clean Ledgers With AI Match, flag and clean ledger entries with AI help and formula checks
- Monthly MIS Reports and Commentary With AI Draft variance commentary for a monthly report and verify every figure
- Use AI in Finance Safely: Checks and Audit Trail Set review steps and a simple log so AI-assisted work passes an audit
AI for HR and OperationsPart of AI Tools for Work (No Code): ChatGPT, Claude, Gemini, Copilot, NotebookLM, Perplexity
- Job Descriptions, Screening and Interviews With AI Write a fair job description, screening rubric and interview questions
- An HR Policy Q&A Assistant From Your Handbook Let staff ask questions of your handbook and get cited answers
- Write SOPs and Automate Ops Checklists Turn a messy process into an SOP and an automated checklist
AI Strategy for Leaders
- AI ROI: Costs, Savings and Real Payback Build a one-page ROI case with honest costs, savings and payback time
- Pick Your First AI Use Cases: A Scoring Grid Score candidate use cases on value, risk, data and effort, and pick three
- Build vs Buy vs Partner for AI Choose build, buy or partner for a use case with a clear checklist
- AI Skills for Your Team: Who Learns What Map the AI skills each role needs and plan training for the quarter
- Roll Out AI: Pilots, Change and Adoption Run a pilot, handle resistance and turn early wins into team habits
- Measure AI Impact: KPIs, Baselines, Reviews Set baselines and KPIs, then run a monthly review that drives decisions
AI for Engineers (Civil, Mechanical, Electrical)
- What AI Can Do for Engineers, and What It Must Not Sort engineering tasks into safe AI help and calculations you must verify yourself
- Specs, Calcs and Reports With AI, Then Verify Summarise a spec, draft a calc sheet and check every number against codes and hand calculations
- Project: Inspection Report and Checklist Pack Build a site or plant inspection checklist and report template with a verification step
AI for IT Professionals (Non-AI Developers)
- AI for IT Pros: Safe Uses Without Building Models Use AI for scripts, queries and docs while keeping secrets and client code out of public tools
- Debug, Script and Document With Free AI Coding Help Debug an error, write a SQL or shell script and document it, testing before you trust it
- Project: Automate One Boring IT Task Safely Automate one repetitive task with AI-written code, tested on dummy data first
AI for Doctors: Admin, Learning and Patient Education
- AI for Doctors: What Is Safe, and What Never Is Know where AI helps with admin and learning, and why it never diagnoses or replaces clinical judgment; patient privacy under DPDP comes first
- Journal Digests and Patient Handouts With AI (No Data) Summarise papers and draft plain-language handouts using no patient identifiers, then verify against guidelines
- Project: A Verified Patient Handout and Journal Digest Produce one patient-education handout and one journal digest, each checked line by line by you
AI for Nurses and Healthcare Staff
- AI for Nurses and Hospital Staff: Safe Uses Only Use AI for training, rosters and communication, never for clinical decisions, with patient data kept out under DPDP
- Shift Handover Templates, Rosters and Training With AI Draft handover templates, duty rosters and training quizzes using no patient identifiers, then have them checked by your seniors
- Project: A Ward Training Kit and Duty Roster Build a training quiz set and a fair duty roster, reviewed against your hospital protocols
AI for Pharmacists and Medical Stores
- AI for Pharmacists: Safe Uses and Hard Limits Use AI for stock and learning tasks while dosage, interaction and dispensing decisions stay with you; customer data stays private under DPDP
- Stock, Expiry Tracking and Counselling Notes With AI Build an expiry tracker and draft plain counselling notes, verified against official drug references
- Project: Expiry Tracker and Reorder Sheet Set up a stock sheet with expiry alerts and reorder suggestions that you verify weekly
AI for Lawyers: Research Aid, Not Legal Advice
- AI for Lawyers: What It Can Do and Where It Fails Use AI for drafting aids and summaries, understand fake-citation risk, and protect client confidentiality under DPDP
- Summarise Judgments and Draft Notices, Then Verify Summarise a judgment and draft a first-pass notice, then verify every citation and clause yourself
- Project: A Verified Case Brief and Client Checklist Build a case brief and client intake checklist where every cited source is checked by you
AI for CAs and Tax Professionals
- AI for CAs and Tax Pros: Safe Uses and Limits Use AI for explanation and drafting while rates, sections and filings are verified against official sources; client data stays private under DPDP
- GST and Tax Notes, Reminders and Client Queries With AI Draft client reminders and plain-language notes on GST and income tax, then verify every rule on official portals
- Project: A Client Compliance Calendar and Query Bank Build a compliance calendar and FAQ bank with each rule checked against the latest official source
AI for Architects
- AI for Architects: Ideas, Not Approvals Use AI for concepts, research and client notes while codes, structure and approvals stay with licensed professionals
- Mood Boards, Briefs and Code Summaries With AI Create a concept brief and mood board and summarise bylaws, verified against the official text
- Project: A Client Concept Presentation Build a client-ready concept deck with AI visuals clearly labelled as concept only
AI for Interior Designers
- AI for Interior Designers: Where It Helps and Misleads Use AI for ideas and client communication while checking real dimensions, materials and costs yourself
- Mood Boards, Material Lists and Quotes With AI Generate mood boards and a material list with a quote sheet, checked against real vendor rates
- Project: A Room Design Pitch for a Client Create a room concept board, material list and quote ready to present
AI for Real Estate Agents
- AI for Real Estate Agents: Safe and Honest Marketing Use AI for listings and follow-ups without false claims, and keep buyer data private under DPDP
- Listings, Lead Follow-Ups and Area Guides With AI Write accurate listings and follow-up messages and an area guide, with every fact checked
- Project: A Listing Kit and Lead Tracker Build a listing description set and a lead tracker with follow-up reminders
AI for Sales Professionals
- AI for Sales: Faster Prep Without Spammy Outreach Use AI to prepare for calls and personalise messages without spam or invented claims
- Prospect Research, Pitch Notes and Follow-Ups With AI Research a prospect, draft a pitch and write follow-ups, with facts checked
- Project: A Pitch Pack and Follow-Up Sequence Build a pitch one-pager, objection answers and a three-step follow-up sequence
AI for Customer Support Teams
- AI for Support Teams: Faster Replies, Human Judgment Use AI to draft replies and tag issues while escalations and customer data stay with people
- Reply Templates, Tagging and Help Articles With AI Turn past tickets into reply templates and help articles, with customer details removed first
- Project: A Support Reply Library and FAQ Page Build a tested reply library and FAQ that your team reviews before use
AI for Journalists and Writers
- AI for Journalists and Writers: Ethics, Facts and Voice Use AI for research aid and editing without inventing facts, quotes or sources, and disclose its use honestly
- Interview Transcripts, Fact-Checks and Edits With AI Transcribe an interview, pull key quotes and fact-check claims against primary sources
- Project: A Researched Article With a Source Log Write a short researched piece with a source log proving each fact
AI for Graphic and UI Designers
- AI for Designers: Ideas, Rights and Originality Use AI for ideation while understanding licence, copyright and originality limits
- Moodboards, Layout Ideas and Copy With Free Design Tools Explore layouts and copy with AI, then refine by hand in Figma or Penpot
- Project: A Brand Kit and Landing Page Mockup Create a small brand kit and landing page mockup, with every asset licence-checked
AI for Content Creators and YouTubers
- AI for Creators: Speed Without Losing Your Voice Use AI for ideas and drafts while keeping your voice, facts and platform rules intact
- Scripts, Titles, Thumbnails and Captions With Free Tools Plan a video from idea to script, title, thumbnail brief and captions with free tools
- Project: A 30-Day Content Calendar Build a content calendar with scripts for the first week and a review checklist
AI for Freelancers
- AI for Freelancers: Do More Without Overpromising Use AI to speed up client work while staying honest about its use and protecting client data
- Proposals, Invoices and Client Updates With AI Draft a proposal, invoice and weekly update, with scope and numbers checked by you
- Project: A Client Onboarding Kit Build a proposal template, scope checklist and invoice tracker for new clients
AI for Shop and Retail Owners
- AI for Shop Owners: Simple Wins and Safe Habits Use AI for offers, messages and stock thinking while keeping customer details private
- Offers, Stock Sheets and WhatsApp Messages With AI Make a stock sheet, festival offers and WhatsApp replies for your shop
- Project: A Festival Sale Plan for Your Shop Plan a festival sale with an offer list, poster and message set
AI for Manufacturing and Factory Managers
- AI for Factory Managers: Safe Uses on the Shop Floor Use AI for reports and planning while safety-critical decisions stay with qualified people
- SOPs, Downtime Logs and Shift Reports With AI Draft SOPs and analyse downtime logs and shift reports, verified on the floor
- Project: A Downtime Tracker and Weekly Plant Report Build a downtime tracker and weekly report that highlights the top three losses
AI for Restaurant and Hospitality Owners
- AI for Restaurant Owners: Safe Uses and Food Safety Limits Use AI for menus, reviews and rosters while food safety and allergen facts are checked by you
- Menus, Review Replies and Staff Rosters With AI Write menu descriptions, reply to reviews and build a staff roster
- Project: A Weekly Menu, Cost Sheet and Review Reply Kit Build a weekly menu with a cost sheet and review reply templates
AI for Farmers and Agri-Business
- AI for Farmers: Helpful Advice, Not Final Advice Use AI for learning and planning while confirming crop, pesticide and market advice with local experts and official sources
- Crop Records, Mandi Prices and Scheme Info With AI Keep crop records, track mandi prices and understand scheme forms, verified on official portals
- Project: A Season Plan and Farm Record Sheet Build a season plan and cost record sheet checked with your local agri office
AI for Government Exam Aspirants
- AI for Exam Aspirants: Study Smarter, Not Shortcuts Use AI as a tutor and planner and stay careful about facts, current affairs and exam rules
- Syllabus Maps, Notes and Daily Practice With AI Turn a syllabus into a study map, notes and daily MCQ practice with answers checked against standard books
- Project: A 90-Day Study Plan and Mock Test Bank Build a 90-day plan and a mock test bank with a weekly review loop