Retrieval & knowledge systems
Generate embeddings and run vector search
Choose embedding models, store vectors (pgvector/Chroma/Pinecone), run similarity and hybrid search.
~10 focused hours·intermediate
Tools: OpenAI/Cohere embeddings, pgvector, Chroma, Pinecone/Qdrant, Hybrid search (BM25 + vector)
Market relevance — share of job ads asking for this
What employers mean
You should be able to…
- Generate embeddings for text chunks and store them in a vector database
- Run similarity search to retrieve the most relevant chunks for a query
- Combine keyword (BM25) and vector search for hybrid retrieval
- Choose an embedding model and dimension appropriate for cost/accuracy tradeoffs
- Set up indexing (HNSW/IVF) for fast approximate nearest-neighbor search at scale
- Re-embed and update the index when source documents change
Needs first: Ingest and chunk documents
Learn — free, link-checked
The few resources that matter
Read · beginner · 30 min · platform.openai.com
Embeddings
Explains embedding models, dimensionality, and similarity metrics with runnable examples. — OpenAI
Read · beginner · 35 min · docs.trychroma.com
Chroma Documentation
Simplest open-source vector DB to get running locally — best on-ramp before touching a managed service. — Chroma
Build from · intermediate · 40 min · github.com
pgvector
The vector extension employers actually ask about when they already run Postgres — indexing (HNSW/IVFFlat) details included. — pgvector (Andrew Kane)
Course · intermediate · 90 min · deeplearning.ai
Building Applications with Vector Databases
Ties embeddings + vector DB choice + hybrid search together into applied examples, not just API syntax. — DeepLearning.AI
Practice
Hybrid search over Indian consumer-tech product reviews
Use a sample dataset of Indian e-commerce product reviews (phones, appliances) and build a hybrid search system: embed reviews into pgvector or Chroma, add a BM25 keyword index, and combine both with re-ranking. Support queries like 'battery drains fast Redmi phone' that need both keyword and semantic matching.
Done when
- At least 500 reviews embedded and indexed in a vector store (pgvector or Chroma)
- A BM25 keyword index runs alongside vector search, with a combined re-ranked result set
- A test set of 10 realistic queries shows hybrid search outperforming vector-only or keyword-only on relevance
- Query latency for top-k retrieval is measured and reported (target: sub-second on the test dataset)
Prove it
Evidence a recruiter can check
- Public GitHub repo with the indexing pipeline and a benchmark table (hybrid vs vector-only vs keyword-only)
- Sample query results pasted in README showing retrieved chunks with relevance
- Latency numbers for retrieval at the tested dataset size
- Notes on embedding model choice and why
Interview
Questions you'll get asked
- How do you choose between pgvector, Chroma, and a managed service like Pinecone for a given project?
- What's the difference between exact and approximate nearest-neighbor search, and when does it matter?
- How would you implement hybrid search combining BM25 keyword matching with vector similarity?
- How do you decide on embedding dimensionality and its cost/latency/accuracy tradeoffs?
- How would you keep a vector index in sync as underlying documents get updated or deleted?
- How do you evaluate whether your embedding model is actually retrieving relevant results?