INTERMEDIATE · 8 MODULES · 64 LESSONS

Advanced RAG Techniques

Take a RAG system past the basics: hybrid search, reranking, query rewriting and evaluation, so your team can see which change actually improves the answers.

  1. 01RAG Pipeline Completo (Architecture Overview)8 LESSONS
    1. Module Introduction: The Complete RAG Pipeline
    2. RAG Pipeline Components
    3. RAG Architecture Decisions
    4. Success Metrics in RAG
    5. Real-World RAG Use Cases in Production
    6. RAG vs Traditional Search (Keywords)
    7. Architecture Overview: Modules 2-8
    8. Project: Baseline RAG System
  2. 02Chunking Strategies8 LESSONS
    1. Introduction to Chunking Strategies
    2. The Problems with Fixed-Size Chunking
    3. Recursive Chunking with LangChain
    4. Capsule 04: Semantic chunking · chunks that respect the change of topic
    5. Capsule 05: Structural chunking · respecting the content's natural units
    6. Capsule 06: Chunk overlap · the insurance against answers cut in half
    7. Capsule 07: The decision framework · which chunking strategy to choose
    8. Project: Chunking Strategy Optimizer
  3. 03Query Optimization8 LESSONS
    1. Introduction to Query Optimization
    2. Capsule 02: Why the user's query is almost never the optimal query
    3. Capsule 03: Query expansion · when one query isn't enough
    4. Capsule 04: Query rewriting · transforming the query before you search
    5. Capsule 05: Query decomposition · when one question is several questions
    6. Capsule 06: HyDE · searching with the hypothetical answer instead of the question
    7. Capsule 07: Decision framework · which query optimization technique to pick
    8. Project: Query Optimizer System
  4. 04Re-ranking Techniques8 LESSONS
    1. Module 4: Re-ranking · the second stage that turns mediocre retrieval into excellent retrieval
    2. Capsule 02: Cosine similarity is good, not perfect · the limit that justifies re-ranking
    3. Capsule 03: Cross-encoder re-ranking · the default option that almost always wins
    4. Capsule 04: LLM-based re-ranking · the expensive option when 3% extra precision justifies the cost
    5. Capsule 05: Cohere Rerank API · the managed option when you don't want to maintain models
    6. Capsule 06: Re-ranking trade-offs and operational optimizations
    7. Capsule 07: Decision framework · which reranker to pick and when
    8. Capsule 08: Capstone project · a re-ranking system with A/B testing
  5. 05Hybrid Search (BM25 + Semantic)8 LESSONS
    1. Module 5: Hybrid Search · combining keyword + semantic for queries that need both
    2. Capsule 02: Why semantic search alone isn't enough · the concrete failure modes
    3. Capsule 03: BM25 · the classic algorithm that still wins for exact keywords
    4. Capsule 04: Reciprocal Rank Fusion · the simple formula that combines rankings from any source
    5. Capsule 05: Weighted hybrid blending · when one signal is clearly better than the other
    6. Capsule 06: Elasticsearch · hybrid search at production scale
    7. Capsule 07: The decision framework · which hybrid search strategy to choose
    8. Capsule 08: Capstone project · a hybrid search engine with A/B testing
  6. 06Metadata Filtering8 LESSONS
    1. Module 6: Metadata Filtering · the component almost nobody implements first but everyone ends up needing
    2. Capsule 02: Pre-filter vs post-filter · the architectural decision that defines the whole module
    3. Capsule 03: Designing the metadata schema · the decision your future self will thank you for (or curse you for)
    4. Capsule 04: Where clauses in ChromaDB · the complete filter syntax
    5. Capsule 05: Multi-tenant isolation · turning isolation into a platform rule
    6. Capsule 06: Time-based and tag filtering · the two most operational filters
    7. Capsule 07: Integration with hybrid search · the final "state of the art" pattern
    8. Capsule 08: Capstone project · an end-to-end Metadata-Filtered RAG
  7. 07Production RAG con Pinecone8 LESSONS
    1. Module 7: Production with Pinecone · the migration from "working demo" to "24/7 service"
    2. Capsule 02: ChromaDB's concrete limits in production
    3. Capsule 03: Setting up serverless Pinecone · your first production-ready index
    4. Capsule 04: The ChromaDB → Pinecone migration · without losing data or breaking queries
    5. Capsule 05: Pinecone namespaces · native multi-tenant isolation, and why it beats metadata filtering
    6. Metadata Filtering in Pinecone
    7. Benchmarking and Costs: ChromaDB vs Pinecone
    8. Project: A Production RAG System with Pinecone
  8. 08RAG Evaluation + Proyecto Integrador8 LESSONS
    1. Module Introduction: RAG Evaluation
    2. RAG Evaluation Fundamentals
    3. Core RAGAS Metrics
    4. The Golden Dataset and Ground Truth for RAG
    5. The Automated Evaluation Pipeline
    6. Regression Testing and Quality Thresholds
    7. Evaluation CI/CD with GitHub Actions
    8. Final Project: An Evaluated Advanced RAG System
Reviews

What students say

These reviews are from enrolled students who completed at least 50% of the course. We moderate reviews only on content grounds (spam, offensive language, personal data), never for being critical or negative.

No approved reviews yet.

Be the first to share your experience!

SEC. 06 · THE STUDIO

Why this studio exists

For decades, building software belonged to technical profiles with years of training, and getting a digital product into the real world took companies months or years. No-code and AI have narrowed that gap, but the knowledge is still held by a few.

Tech Benders exists to open that knowledge up, so business teams and non-technical teams can put their ideas into production with the right technical decisions behind them. AI does what you ask of it. Without a technical base, what it builds may not scale with your company’s ecosystem, and its costs can run away over time.

The team knows the pains of large organizations and the speed at which they have to move. It brings solution architecture, data engineering, product built end to end from back end to front end, and daily work with AI. Software that solves what you need in days, not months, with the cost of AI kept in check.

Founders

The studioRead about Tech Benders →

Write to ushi@techbenders.io

Made with Wislab