Master RAG engineering
Build production-ready RAG systems from fundamentals to cutting-edge research. Learn through interactive labs and real-world challenges.
68
Lessons
268+
Challenges
81
Modules
15
Phases
13-Phase Curriculum
From vector math to production ops.
Interactive Labs
Write code, run tests, instant feedback.
Decision Playbooks
Matrices for models, LLMs, guardrails.
Project Certificates
Earn certificates for completed tracks.
Popular Challenges
Community favorites to get started
Advanced RAG Architectures
Master cutting-edge techniques from the latest research papers.
MiA-RAG
Mindscape-Aware Context
Hierarchical summarization for global document understanding
QuCo-RAG
Corpus-Based Uncertainty
Trigger retrieval using pre-training corpus statistics
HiFi-RAG
Hierarchical Filtering
Multi-stage filtering for maximum context precision
Graph-O1
MCTS Reasoning
Monte Carlo Tree Search for graph exploration
What's New
Recent additions to the platform
Voice RAG
Phase 7Audio-based retrieval and generation
MCP Integration
Phase 5Model Context Protocol support
Project Certificates
ProjectsEarn certificates for completed tracks
Peer Reviews
ProjectsCommunity feedback on submissions
Your Learning Journey
13 phases from fundamentals to production
Foundations
12
Pre-Retrieval
48
Retrieval
19
Query Transforms
15
Advanced Retrieval
10
Post-Retrieval
20
Grounding & Safety
19
Agentic RAG
25
Graph RAG
8
Multimodal
11
Fine-tuning
7
Production Ops
30
Evaluation Ops
38
Hands-On Projects
Build production systems, earn certificates
A production RAG system, step by step.
Scroll through the pipeline to understand how each component works together.
Ingest with traceability
Parse → chunk → attach metadata → stable chunk IDs. Debuggability beats cleverness.
- Stable chunk_id + doc_id + offsets
- Structure-aware chunking (headers, tables, code)
- Dedup + versioning so re-indexing is safe
Maximize recall (hybrid + fusion)
Default production strategy: BM25 + dense embeddings fused with RRF.
- Metadata filters first (tenant/doc_type/date)
- Hybrid retrieval (BM25 + dense)
- RRF fusion to avoid score-normalization pitfalls
Maximize precision (rerank)
Rerank top‑20/50 candidates with a cross‑encoder, keep the best 5–10.
- Reranker cascade: cheap shortlist → expensive rerank
- MMR diversity to reduce redundancy
- Lost-in-the-middle ordering
Ground answers (and refuse safely)
Citations, refusal policies, injection/PII defenses — production hardening isn't optional.
- Citation validation (no out-of-range cites)
- Refuse on insufficient context
- Sanitize prompt injection + redact PII before LLM
RAG pipeline (scrollytelling)
Scroll left side — diagram reacts as pipeline moves from ingestion to retrieval, reranking, and safety.
Build production RAG systems
Master retrieval, grounding, agents, and evaluation through 260+ interactive challenges. Earn certificates. Ship with confidence.