INTERMEDIATE · 8 MODULES · 64 LESSONS

Production RAG and Document Ingestion

Give your agent reliable access to real documents: clean them, chunk them with care, search them like any other tool, and cite the source in every answer.

For you if
You have an agent with tools and need it to answer from your policies, manuals, or FAQs.
By the end
You'll know how to process messy documents, measure retrieval quality, and answer without making things up when nothing matches.
You need
Intermediate Python and an agent with tool calling under your belt. No API key needed.
  1. 01Parsing and Chunking Documents8 LESSONS
    1. Module 1: Parsing and Chunking Documents
    2. Why Ingestion Is the Biggest Block
    3. Parsing HTML with `html.parser`
    4. Parsing Markdown and Cleaning Dirty Text
    5. Chunking Strategies: Fixed Size vs. Structure-Aware
    6. The Chunk-Size Trade-off
    7. Metadata for Citation
    8. Mini-project: Ingest the Reservo Corpus
  2. 02Indexing Chunks for Retrieval8 LESSONS
    1. Module 2: Indexing Chunks for Retrieval
    2. What Is a Retrieval Index
    3. The Inverted Index
    4. Term Frequency and IDF
    5. BM25 Scoring
    6. The Lexical Limit: A Synonym That Misses
    7. search(query, k), Executed
    8. Mini-project: A BM25 Index over Reservo
  3. 03`search_docs` as an Agent Tool8 LESSONS
    1. Module 3: `search_docs` as an Agent Tool
    2. From Index to Tool
    3. The `search_docs` Contract
    4. What the Tool Returns to the Model
    5. Writing the `description` So the Model Knows When
    6. Bounding the Tool: `k` and Truncation
    7. `search_docs` Executed End to End
    8. Mini-project: `search_docs` as One More Tool
  4. 04Agentic Retrieval in the Loop8 LESSONS
    1. Module 4: Agentic Retrieval in the Loop
    2. When Does the Agent Search
    3. Reformulating and Retrying the Query
    4. Multi-hop Retrieval
    5. Combining `search_docs` with Reservo's Tools
    6. Grounding the Answer in Chunks
    7. Agentic Retrieval vs. Monolithic RAG
    8. Mini-project: An Agentic Retrieval Run
  5. 05Incremental and Idempotent Ingestion8 LESSONS
    1. Module 5: Incremental and Idempotent Ingestion
    2. Documents change in production
    3. Content hashing for idempotency
    4. Detecting new, modified, and deleted documents
    5. Reingesting without duplicating
    6. Updating only what changed
    7. Handling a deleted document
    8. Mini-project: an incremental ingest
  6. 06Evaluating Retrieval Quality8 LESSONS
    1. Module 6: Evaluating Retrieval Quality
    2. Why evaluate retrieval, not generation
    3. The fixed evaluation set
    4. Recall@k
    5. Precision@k
    6. When lexical retrieval fails
    7. The path to better recall
    8. Mini-project: a retrieval evaluation harness
  7. 07Operating RAG in Production8 LESSONS
    1. Module 7: Operating RAG in Production
    2. Citing sources in the answer
    3. Handling zero results
    4. Filtering irrelevant chunks with a threshold
    5. A source document disappeared
    6. Measuring what gets retrieved
    7. The cost of the pipeline
    8. Mini-project: operating `search_docs`
  8. 08Module 8 (project): A Production-Ready `search_docs`8 LESSONS
    1. Module 8 (project): A Production-Ready `search_docs`
    2. Assembling the ingestion pipeline
    3. The index and the tool
    4. Agentic retrieval end to end
    5. Incremental reingest in the capstone
    6. The evaluation score
    7. What your RAG still needs
    8. Project: shipping `search_docs`
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