Zero to GenAI Engineer

Session 00 — Search engines (PDF)

• GenAI introWhat is GenAI
• How search engines workCrawling → TF-IDF
• Claude Code leakHow LLMs are instructed

Session 01 — Text to numbers (PDF)

• Text to NumbersBoW → FastText

Session 02 — Transformers (PDF)

• Transformer architectureAttention, encoder–decoder, positional encoding

Session 03 — GPT papers

• GPT-1 → DPO paper walkthrough14 slides · 11 papers

Session 05 — Local LLMs (PDF)

• Ollama · LM Studio · OpenRouterSession PDF

Session 08 — Recap

• Interactive recap of S00–S07Visual slides
• Full-text recap slidesReadable version

Session 10 — RAG

• 01 — Why RAG
• 02 — Chunking (LangChain)
• 03 — Chunking (LlamaIndex)
• 04 — Embeddings
• 05 — Vector databases
• Revision — notebooks 01–05
• 06 — Why BM25
• 07 — Hybrid search
• 08 — Reranking
• 09 — Full pipeline recap
• 11 — Production chatbots (memory)
• 15 — Multimodal RAG
• RAG Studio — strategy evaluation report

Session 11 — LangGraph

• 01 — Fundamentals & agents
• 02 — Human-in-the-loop
• 04 — Reasoning patterns
• 05 — SQL agent

Session 12 — Deep Agents

• 01 — Why Deep AgentsLLM vs agent vs harness · AGENT.md / SKILL.md

Session 15 — LLM Fine-Tuning

• 01 — Fine-tuning · LoRA · QLoRARAG vs fine-tune · PEFT · quantization · Unsloth path

Session 17 — Decision Models

• 01 — RLCD · the training behind JevCalibration · Brier score · parallel decisions · threshold gating

Session 18 — Final project

• 01 — AI Talking Avatar · expected demoFour-beat kiosk: C22 · refuse NIA 442 · photo face · Hindi

Full-course recap · S00 to S15

• Interactive recap · S00 to S15Dotted timeline · interview Q&A · does not replace Session 08