When gemini-embedding-2 Retrieval Feels 'Almost Right,' Check task_type First
When gemini-embedding-2 search misses in a frustrating near-hit way, the cause is often a missing or mismatched task_type. Here is how to align document and query intent, plus the code and a tiny harness to prove the difference on your own data.
When Your Knowledge Base Shifts Mid-Run: Pinning File Search to an Execution Epoch for Consistent Agent Grounding
When a File Search store is updated while a Managed Agent is running, a single execution can mix old and new grounding. Borrowing MVCC ideas, pinning an execution epoch keeps one agent run's evidence consistent. Here is the design and implementation.
One File Search Store for Many Apps: Splitting Retrieval With customMetadata and Chunk Config
Put several apps' FAQs in a single Gemini File Search store and metadataFilter can silently return empty grounding, or answers get split across chunk boundaries. Here is the customMetadata design, the AIP-160 filter-syntax trap, and measured chunkingConfig tuning.
Ingested but Never Cited: Pruning a File Search Store with Citation Logs and a Quarantine Window
Most documents in a File Search store are never cited, quietly draining both cost and retrieval quality. Learn to log grounding metadata, surface never-cited documents from real usage data, and prune them safely with a quarantine window — with working code.
When Gemini × Qdrant Hybrid Search Was Quietly Losing Recall — Field Notes on Instrumenting RRF Weights and Sparse-Vector Drift
Run Gemini embeddings with Qdrant hybrid search in production and your dashboards stay green while recall quietly slips. These field notes show how to catch it with measurement — RRF weights, sparse-vector drift, missing payload indexes — and protect it with a quality budget.
Citing the exact page and figure in File Search answers with visual-citation metadata
File Search grounding metadata now carries media_id and page_numbers, so you can trace each sentence of an answer back to a specific page and figure. Here's how I built a sentence-level, verifiable citation layer over a mix of PDFs and images.
Your File Search Store Goes Stale in Production — Catalog Sync and Drift Detection That Actually Hold
Load a catalog into File Search once and forget it, and within weeks it starts confidently pointing users at assets you already pulled. Here is the sync pipeline I run: hash-based incremental import, a blue/green rebuild that swallows deletions, and a nightly drift audit.
Put Help Docs and Screenshots in One File Search Store and Return Answers That Cite the Image Too
Your text help docs and your screenshots live in separate stores, so a single question can never return both the steps and the matching screen. With gemini-embedding-2 going multimodal in File Search, here is how I merged them and returned the cited screenshot alongside the answer.
When Your Firestore × Gemini Embeddings RAG Quietly Degrades — Designing for Re-Embedding
A RAG built on Firestore native vector search and Gemini Embeddings drifts when the embedding model changes generations, and retrieval quality drops with no errors. Here is how to detect the drift, re-embed without downtime, and keep retrieval cost in check.
Permission-Aware RAG — Designing Gemini Search That Only Cites What the User Is Allowed to See
The day you add RAG to internal search, drafts and finance memos nobody should see start leaking into answers. This is a production design — metadata filtering, defense in depth, and audit logging — for letting Gemini search while respecting permissions, with working code.
Rebuilding a Three-Layer RAG Cache After Migrating to Gemini 3.5 Flash
When Gemini 2.0 Flash was retired, I rebuilt my RAG caching stack around 3.5 Flash. Here are the working implementations for response, semantic, and embedding caches, measured hit rates from production, and how self-managed caching divides the work with the API's Context Caching.
Trimming Gemini Embeddings from 3072 to 768 Dimensions: A Matryoshka Approach to Cutting Vector DB Cost and Latency
gemini-embedding-001 returns 3072-dimensional vectors, but thanks to Matryoshka representation you can keep only the leading dimensions with almost no quality loss. This is a design for trimming to 768 to cut vector DB storage and latency, including the re-normalization pitfall and coarse-to-fine search code.