Build your plan
Pick a pace and we distribute every remaining roadmap resource across your week and generate a calendar you can import anywhere.
1. Set your pace
Hours per week:
75
resources to cover
48h
timed content
7
weeks needed
Your week-by-week schedule
Week 1 (Oct 5 – Oct 11)
4.6h- Oct 8Big-picture: what an LLM is · P1 · 8m
- Oct 8Full forward pass, embeddings, softmax · P1 · 27m
- Oct 8Attention: Q/K/V, masking, multi-head · P1 · 26m
- Oct 8How facts live in MLP layers (often skipped) · P1 · 8m
- Oct 9How facts live in MLP layers (often skipped) · P1 · 15m
- Oct 9Tokenization/BPE + positional encoding · P1 · 53m
- Oct 10Tokenization/BPE + positional encoding · P1 · 63m
- Oct 10APIs, prompts, streaming, multi-provider · P2 · 6m
- Oct 11APIs, prompts, streaming, multi-provider · P2 · 14m
- Oct 11Structured output (JSON/Pydantic) · P2 · 20m
- Oct 11Why RAG; embeddings; vector DB; cosine · P3 · 34m
Week 2 (Oct 12 – Oct 18)
8h- Oct 12Why RAG; embeddings; vector DB; cosine · P3 · 4m
- Oct 12Build a real RAG app · P3 · 20m
- Oct 12Hybrid search: BM25 + dense + RRF · P3 · 45m
- Oct 13Hybrid search: BM25 + dense + RRF · P3 · 69m
- Oct 14Hybrid search: BM25 + dense + RRF · P3 · 69m
- Oct 15Hybrid search: BM25 + dense + RRF · P3 · 69m
- Oct 16Hybrid search: BM25 + dense + RRF · P3 · 69m
- Oct 17Hybrid search: BM25 + dense + RRF · P3 · 69m
- Oct 18Hybrid search: BM25 + dense + RRF · P3 · 69m
Week 3 (Oct 19 – Oct 25)
8h- Oct 19Hybrid search: BM25 + dense + RRF · P3 · 2m
- Oct 19Reranking (cross-encoders/ColBERT) · P3 · 9m
- Oct 19Query transforms (HyDE, multi-query) · P3 · 20m
- Oct 19Chunking: parent-child, sentence-window · P3 · 20m
- Oct 19GraphRAG / RAPTOR / Contextual Retrieval · P3 · 18m
- Oct 20GraphRAG / RAPTOR / Contextual Retrieval · P3 · 2m
- Oct 20Metrics: faithfulness, context precision/recall · P3 · 12m
- Oct 20RAGAS hands-on in Python · P3 · 6m
- Oct 20The one rule: fine-tune = behavior, not knowledge · P4 · 20m
- Oct 20LoRA/QLoRA + rank, alpha, dropout, 4-bit NF4 · P4 · 15m
- Oct 20Quick PEFT overview (LoRA/QLoRA/DoRA) · P4 · 4m
- Oct 20End-to-end deep dive + code · P4 · 9m
- Oct 21End-to-end deep dive + code · P4 · 69m
- Oct 22End-to-end deep dive + code · P4 · 9m
- Oct 22Agent vs workflow; ReAct loop · P5 · 10m
- Oct 22ReAct in practice · P5 · 6m
- Oct 224 patterns + evals (whole course) · P5 · 44m
- Oct 234 patterns + evals (whole course) · P5 · 69m
- Oct 244 patterns + evals (whole course) · P5 · 69m
- Oct 254 patterns + evals (whole course) · P5 · 69m
Week 4 (Oct 26 – Nov 1)
8h- Oct 264 patterns + evals (whole course) · P5 · 69m
- Oct 274 patterns + evals (whole course) · P5 · 69m
- Oct 284 patterns + evals (whole course) · P5 · 69m
- Oct 294 patterns + evals (whole course) · P5 · 10m
- Oct 29Reflection / Tool use / Planning / Multi-agent · P5 · 20m
- Oct 29Build 8 agent projects · P5 · 20m
- Oct 29Context window as scarce RAM; write/select/compress/isolate · P6 · 14m
- Oct 29Anthropic's mental model + failure modes · P6 · 5m
- Oct 30Anthropic's mental model + failure modes · P6 · 15m
- Oct 30Short vs long vs episodic memory · P6 · 22m
- Oct 30Memory taxonomy + tools (Mem0, Letta, Zep) · P6 · 8m
- Oct 30LangChain vs LangGraph vs LangSmith · P7 · 12m
- Oct 30LangGraph: state, nodes, edges, memory, HITL · P7 · 11m
- Oct 31LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
- Nov 1LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
Week 5 (Nov 2 – Nov 8)
8h- Nov 2LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
- Nov 3LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
- Nov 4LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
- Nov 5LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
- Nov 6LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
- Nov 7LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
- Nov 8LangGraph: state, nodes, edges, memory, HITL · P7 · 69m
Week 6 (Nov 9 – Nov 15)
8h- Nov 9LangGraph: state, nodes, edges, memory, HITL · P7 · 32m
- Nov 9LangGraph full free course · P7 · 37m
- Nov 10LangGraph full free course · P7 · 69m
- Nov 11LangGraph full free course · P7 · 69m
- Nov 12LangGraph full free course · P7 · 6m
- Nov 12CrewAI / OpenAI SDK / AutoGen / ADK · P7 · 20m
- Nov 12Tracing & debugging (LangSmith) · P7 · 20m
- Nov 12Why MCP (2-min mental model) · P8 · 8m
- Nov 12Why → Architecture → Lifecycle → Build · P8 · 15m
- Nov 13Why → Architecture → Lifecycle → Build · P8 · 5m
- Nov 13Full build: server+client, tools, resources, prompts, SAMPLING · P8 · 63m
- Nov 14Full build: server+client, tools, resources, prompts, SAMPLING · P8 · 12m
- Nov 14Production server w/ FastMCP + auth · P8 · 57m
- Nov 15Production server w/ FastMCP + auth · P8 · 45m
- Nov 15Hands-on labs (multi-language) · P8 · 20m
- Nov 15Tool poisoning, line jumping, rug pulls · P8 · 3m
Week 7 (Nov 16 – Nov 22)
3.4h- Nov 16Tool poisoning, line jumping, rug pulls · P8 · 17m
- Nov 16Full threat model + defenses · P8 · 20m
- Nov 16Auth, credential risk, CVEs · P8 · 20m
- Nov 16Objective vs LLM-judge; error analysis; traces · P9 · 12m
- Nov 17Objective vs LLM-judge; error analysis; traces · P9 · 8m
- Nov 17RAG evals (RAGAS): faithfulness, precision, recall · P9 · 12m
- Nov 17RAGAS in code (run in CI) · P9 · 20m
- Nov 17Tracing/observability in practice · P9 · 20m
- Nov 17All 10 risks in 4 minutes (overview) · P10 · 4m
- Nov 17LLM01 Prompt Injection (direct/indirect) · P10 · 4m
- Nov 18LLM01 Prompt Injection (direct/indirect) · P10 · 21m
- Nov 18Deep dive w/ real exploits & fixes · P10 · 30m
- Nov 18Practitioner guide (every entry + controls) · P10 · 18m
- Nov 19Practitioner guide (every entry + controls) · P10 · 0m