5modules started4 with hands-on work done
Where I am
- Foundations10/10
- Models & APIs7/7
- Apps without frameworks5/5
- LangChain4/6
- Automation1/4
Next up: M5 · RAG
4module projects builtdetected from the code
Projects detected
- FoundationsM1
- Models & APIsM2
- Apps without frameworksM3
- LangChainM4
29commitson 9 active days · best streak 2 days
Activity
- Active days9
- Best streak2 days
- Current streak1 day
- Started27 Sep 2026
- Latest20 Oct 2026
2,545lines of code54 Python files · 2 notebooks
Lines by project
- Call Transcript Qa801
- Module0729
- Finetuning Project515
- API vs Local175
- Prompts In Langchain118
13practice assignmentssolved
Latest practice
- sets#21
- dictionaries#20
- tuples#19
- functions#18
- docstrings#17
19lecture notes23,313 words · cover 9 theory topics
Notes by week
- Week 15 lectures
- Week 21 lecture
- Week 32 lectures
- Week 411 lectures
Latest: Study resources (Week 4)
Curriculum roadmap
The program's 10 modules and capstone, from the brochure. The whole repo is read on every push: a topic is ticked when my code uses it, or, for theory, when my lecture notes cover it. Every tick links to the file and line that proves it. "Hands-on done" means the module's project is built and most of its topics are covered.
All modules at a glance
| Module | Schedule | Status | Topics | Project |
|---|---|---|---|---|
| M1Foundations of Generative AIWeeks 1–2 | Weeks 1–2 | Hands-on done | 10/10 | ✓ Built |
| M2Working with Models & APIs NowWeeks 3–4 | Weeks 3–4 | Hands-on done | 7/7 | ✓ Built |
| M3Building GenAI Apps Without FrameworksWeek 5 | Week 5 | Hands-on done | 5/5 | ✓ Built |
| M4LangChain Essentials & Application DevelopmentWeeks 6–7 | Weeks 6–7 | Hands-on done | 4/6 | ✓ Built |
| M5RAG Systems — From Basics to AdvancedWeeks 8–9 | Weeks 8–9 | Not started | 0/9 | — |
| M6Agents, Tools & MCP (Model Context Protocol)Weeks 10–11 | Weeks 10–11 | Not started | 0/6 | — |
| M7Agentic Workflows with LangGraph + Final CapstoneWeek 12 | Week 12 | Not started | 0/5 | — |
| M8Practical GenAI Tooling & Productivity EcosystemMonth 4 | Month 4 | Not started | 0/4 | — |
| M9Agentic Ecosystems, Automation Workflows & Applied GenAIMonth 5 | Month 5 | In progress | 1/4 | — |
| M10Career Acceleration, Industry Readiness & Portfolio BuildingMonth 6 | Month 6 | Not started | 0/4 | — |
| ★Capstone Project2 weeks | 2 weeks | Not started | 0/4 | — |
Topics in each module
Foundations of Generative AI
Weeks 1–2 · 10/10 topics
| Topic | Covered by | Proof |
|---|---|---|
| Python refresher | Code | 10_strings_practice.py +20 more |
| Jupyter notebooks | Code | script2-qlora-finetune.ipynb +1 more |
| Hugging Face Hub | Code | script1_baseline_chat.py +6 more |
| Tokens, embeddings and context windows | Notes | lecture 3 +1 more |
| Responsible AI: bias, harm, safety, guardrails | Notes | lecture 4 |
| Roadmap for developers in GenAI | Notes | lecture 0 |
| Evolution of AI: ML → deep learning → GenAI | Notes | lecture 3 |
| Traditional AI vs generative AI | Notes | lecture 1 +1 more |
| Transformers explained (self-attention) | Notes | lecture 3 |
| Pre-trained vs fine-tuned LLMs | Notes | lecture 3 |
| Project First GenAI-powered Python script that sends prompts and parses responses compare_models.py | ||
Working with Models & APIs
Weeks 3–4 · 7/7 topics
| Topic | Covered by | Proof |
|---|---|---|
| Cloud model APIs (OpenAI, Anthropic, Gemini, Mistral) | Code | compare_models.py +5 more |
| Hugging Face models (Inference API / Transformers) | Code | compare_models.py +6 more |
| Local LLMs (Ollama, LM Studio, vLLM) | Code | compare_models.py +1 more |
| Cost management, prompt budgeting and rate limits | Code | compare_models.py +9 more |
| Comparing model behaviour across providers | Code | compare_models.py +2 more |
| Prompt tuning vs fine-tuning, LoRA / QLoRA | Code | script2-qlora-finetune.ipynb +7 more |
| Developer mindset: we orchestrate, not train | Notes | lecture 5 +2 more |
| Project Build and compare a simple chatbot across API, open-source and local models compare_models.py | ||
Building GenAI Apps Without Frameworks
Week 5 · 5/5 topics
| Topic | Covered by | Proof |
|---|---|---|
| Production-style prompt pipelines written by hand | Code | compare_models.py +1 more |
| Conversation memory without frameworks | Code | app.py |
| Parsing LLM responses reliably (JSON, structured parsing) | Code | experiment.ipynb +3 more |
| Simple UI with Streamlit or Gradio | Code | app.py |
| Error handling, retries and fallbacks | Code | compare_models.py +1 more |
| Project Minimal chatbot with memory and a UI, using only Python and APIs (no LangChain) app.py | ||
LangChain Essentials & Application Development
Weeks 6–7 · 4/6 topics
| Topic | Covered by | Proof |
|---|---|---|
| LangChain components: LLMs, prompts, chains, tools, agents | Code | llm_factory.py +9 more |
| Prompt templates and few-shot prompting | Code | 01_basic_prompt_template.py +6 more |
| Structured output and output parsers | Code | experiment.ipynb +3 more |
| Sequential and router chains (LCEL runnables) | Code | experiment.ipynb +2 more |
| Memory in LangChain | — | Not yet |
| LangSmith tracing (optional) | — | Not yet |
| Project Multi-step content generation pipeline (e.g. rewrite → summarize → social snippet) experiment.ipynb | ||
RAG Systems — From Basics to Advanced
Weeks 8–9 · 0/9 topics
| Topic | Covered by | Proof |
|---|---|---|
| Document loaders, text splitters and chunking strategy | — | Not yet |
| Embeddings and vector stores (FAISS, Chroma) | — | Not yet |
| Retrievers (multi-query, contextual compression) | — | Not yet |
| RAG evaluation | — | Not yet |
| Query rewriting and HyDE | — | Not yet |
| Hybrid search | — | Not yet |
| Two-stage retrieval and reranking | — | Not yet |
| Multi-source retrieval and fusion | — | Not yet |
| Context distillation and hallucination mitigation | — | Not yet |
| Project End-to-end "Chat with your documents" assistant with citations | ||
Agents, Tools & MCP (Model Context Protocol)
Weeks 10–11 · 0/6 topics
| Topic | Covered by | Proof |
|---|---|---|
| Tool calling: functions, APIs, custom Python tools | — | Not yet |
| Building multi-tool agents | — | Not yet |
| Safe and predictable agent behaviour | — | Not yet |
| MCP fundamentals and building MCP servers | — | Not yet |
| Using MCP inside agent workflows | — | Not yet |
| Why agents: where they shine and where they don't | — | Not yet |
| Project Multi-tool agent that uses both LangChain tools and MCP tools | ||
Agentic Workflows with LangGraph + Final Capstone
Week 12 · 0/5 topics
| Topic | Covered by | Proof |
|---|---|---|
| LangGraph basics | — | Not yet |
| Nodes, edges and state | — | Not yet |
| Conditional, parallel and iterative workflows | — | Not yet |
| Using tools inside LangGraph nodes | — | Not yet |
| Streaming and persistence (checkpoints) | — | Not yet |
| Project Small LangGraph workflow (e.g. plan → retrieve → summarize with checks) | ||
Practical GenAI Tooling & Productivity Ecosystem
Month 4 · 0/4 topics
| Topic | Covered by | Proof |
|---|---|---|
| AI-first coding workflows (Cursor, Windsurf, Claude Code) | — | Not yet |
| Running local GGUF models with LM Studio | — | Not yet |
| MCP-based productivity agents | — | Not yet |
| Replit Agents | — | Not yet |
| Project Building and deploying a real GenAI project with a mentor | ||
Agentic Ecosystems, Automation Workflows & Applied GenAI
Month 5 · 1/4 topics
| Topic | Covered by | Proof |
|---|---|---|
| Multi-agent orchestration with CrewAI | — | Not yet |
| Automation with n8n | — | Not yet |
| Workflow automation with MCP servers | — | Not yet |
| Microsoft Copilot stack | Notes | study_resources.md |
| Project Domain-specific automation project (finance, marketing, ops) | ||
Career Acceleration, Industry Readiness & Portfolio Building
Month 6 · 0/4 topics
| Topic | Covered by | Proof |
|---|---|---|
| GenAI portfolio on GitHub | — | Not yet |
| Resume building for GenAI roles | — | Not yet |
| Interview preparation for AI engineering roles | — | Not yet |
| Building a personal learning system | — | Not yet |
| Project Industry-ready portfolio of GenAI projects | ||
Capstone Project
2 weeks · 0/4 topics
| Topic | Covered by | Proof |
|---|---|---|
| LLMs (cloud and local) | — | Not yet |
| RAG | — | Not yet |
| Agents and MCP | — | Not yet |
| A full UI | — | Not yet |
| Project Portfolio-ready GenAI app combining LLMs, RAG, agents, MCP, LangGraph and a full UI | ||
Coding activity
Commits per day to the course repo, across the program's 26 weeks.
FewerMore
1 · 2–3 · 4–6 · 7+ commits
Show as a table (commits per week)
| Week starting | Commits |
|---|---|
| 21 Sep 2026 | 3 |
| 28 Sep 2026 | 11 |
| 5 Oct 2026 | 5 |
| 12 Oct 2026 | 7 |
| 19 Oct 2026 | 3 |
What I've built
Every folder in the course repo with its own requirements.txt.
Python foundations
01_module0/
Lessons and practice assignments covering Python fundamentals, from print() to sets.
standard library only
Cloud APIs vs local models
02_module1/week_02/01_API_VS_local/
The same prompts through OpenAI, Gemini, Hugging Face and a local Ollama model, compared side by side.
dotenvgoogle.generativeaiollamaopenairequestsCLI chatbot with switchable providers
02_module1/week_02/02_chatbot/
Command-line chatbot whose provider and model come from a config file.
dotenvgoogle.generativeaiopenaiStreamlit chat assistant with memory
02_module1/week_02/03_capstone/
Web chat UI with conversation history, routing to OpenAI or Gemini from config.
dotenvgoogle.generativeaiopenaistreamlitFinetuning Project
02_module1/week_04/Finetuning-Project/
datasetsjsonlinespefttorchtransformersLLM In Langchain
03_module2/week_01/01_llm_in_langchain/
dotenvlangchain_google_genailangchain_openaiPrompts In Langchain
03_module2/week_01/02_prompts-in-langchain/
dotenvlangchain_corelangchain_google_genailangchain_openaiCall-transcript classification and QA evaluation
99_live_sessions/01_call_transcript_qa/
Live-session project. A LangChain pipeline that classifies customer-support calls and scores tone, resolution and knowledge accuracy with structured outputs.
dotenvfastapilangchain_corelangchain_google_genailangchain_openainumpypandaspydanticsrctqdmLibraries I've used
Third-party imports across all my code: number of files that use each.
dotenv14langchain_core7langchain_google_genai7langchain_openai7pandas6src6google.generativeai5torch5transformers5openai4peft4pydantic4datasets3jsonlines3ollama2requests2tqdm2fastapi1numpy1streamlit1
Show as a table
| Library | Files |
|---|---|
| dotenv | 14 |
| langchain_core | 7 |
| langchain_google_genai | 7 |
| langchain_openai | 7 |
| pandas | 6 |
| src | 6 |
| google.generativeai | 5 |
| torch | 5 |
| transformers | 5 |
| openai | 4 |
| peft | 4 |
| pydantic | 4 |
| datasets | 3 |
| jsonlines | 3 |
| ollama | 2 |
| requests | 2 |
| tqdm | 2 |
| fastapi | 1 |
| numpy | 1 |
| streamlit | 1 |
Recent commits
- Add prompts in LangChain: basic, chat and few-shot prompt templates20 Oct 2026 ·
85a5d83 - Update links in notes and READMEs to the new folder names20 Oct 2026 ·
898b8e5 - Reorganize course folders by module20 Oct 2026 ·
2a59655 - Add FastAPI app, main.py entry point and Docker setup for transcript QA18 Oct 2026 ·
7cad701 - Correct the float16 explanation in the fine-tuning issues log18 Oct 2026 ·
695f52f - add README18 Oct 2026 ·
bba276d - add qlora script that i ran on kaggle.18 Oct 2026 ·
41c6d4d - add notes from the slides provided in week 4.17 Oct 2026 ·
6d246f2 - add notes for week 4 and update README files.17 Oct 2026 ·
6758fd7 - Finetune mini-project17 Oct 2026 ·
9a4b7d1