AutoGineer
Generated banner for Vector Index Tuning
Generated logo for Vector Index Tuning

Vector Index Tuning

Seth HobsonCommunityCreated by Seth Hobson

  • Skill
  • Guidelines
  • Data and AI
  • MIT
  • No issues found
Download ZIP

From the author

Optimize vector index performance for latency, recall, and memory. Use when tuning HNSW parameters, selecting quantization strategies, or scaling vector search infrastructure.
— the description by Seth Hobson

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

Vector Index Tuning provides a structured approach to optimizing vector search infrastructure. It covers index selection based on data scale, HNSW parameter configuration, and various quantization strategies to balance memory footprint with search accuracy. The skill includes best practices for benchmarking, monitoring, and maintaining production vector indexes.

What it does

  • Provides criteria for selecting index types based on dataset size.
  • Explains the impact of HNSW parameters like M, efConstruction, and efSearch.
  • Outlines quantization methods including FP16, INT8, and Product Quantization.
  • Offers templates for benchmarking HNSW configurations.
  • Includes code for implementing scalar quantization.
  • Defines best practices for production monitoring and maintenance.

When to use it

  • When tuning HNSW parameters for specific recall or latency targets.
  • When selecting a quantization strategy to optimize memory usage.
  • When scaling vector search infrastructure to handle millions of vectors.
  • When troubleshooting performance bottlenecks in vector search.
  • When planning for index maintenance and reindexing cycles.

Advantages

  • Helps balance the trade-off between search speed and accuracy.
  • Provides clear guidance on memory optimization through quantization.
  • Includes actionable templates for benchmarking and parameter recommendation.
  • Encourages data-driven tuning using real query benchmarks.

Usage examples

“How should I choose an index type for 50 million vectors?”
The skill recommends using IVF + PQ or DiskANN for datasets exceeding 100 million vectors, suggesting that for 50 million, these advanced structures are appropriate to manage memory and latency.
“What are the recommended HNSW parameters for high recall?”
The skill explains that increasing M and efSearch improves recall, and provides a template function to calculate optimal parameters based on target recall and latency requirements.

Install

  1. Download the skill as a ZIP:

    https://autogineer.com/agent-skills/wshobson/vector-index-tuning/download

  2. Unzip it into the cross-client skills folder of your project:

    mkdir -p .agents/skills && unzip vector-index-tuning.zip -d .agents/skills

Documentation by Seth Hobson

Vector Index Tuning

Guide to optimizing vector indexes for production performance.

When to Use This Skill

  • Tuning HNSW parameters

  • Implementing quantization

  • Optimizing memory usage

  • Reducing search latency

  • Balancing recall vs speed

  • Scaling to billions of vectors

Core Concepts

1. Index Type Selection
Data Size           Recommended Index
────────────────────────────────────────
< 10K vectors  →    Flat (exact search)
10K - 1M       →    HNSW
1M - 100M      →    HNSW + Quantization
> 100M         →    IVF + PQ or DiskANN
2. HNSW Parameters
ParameterDefaultEffect
M16Connections per node, ↑ = better recall, more memory
efConstruction100Build quality, ↑ = better index, slower build
efSearch50Search quality, ↑ = better recall, slower search
3. Quantization Types
Full Precision (FP32): 4 bytes × dimensions
Half Precision (FP16): 2 bytes × dimensions
INT8 Scalar:           1 byte × dimensions
Product Quantization:  ~32-64 bytes total
Binary:                dimensions/8 bytes

Templates and detailed worked examples

Full template library and detailed worked examples live in references/details.md. Read that file when you need the concrete templates.

Best Practices

Do's

  • Benchmark with real queries - Synthetic may not represent production

  • Monitor recall continuously - Can degrade with data drift

  • Start with defaults - Tune only when needed

  • Use quantization - Significant memory savings

  • Consider tiered storage - Hot/cold data separation

Don'ts

  • Don't over-optimize early - Profile first

  • Don't ignore build time - Index updates have cost

  • Don't forget reindexing - Plan for maintenance

  • Don't skip warming - Cold indexes are slow

Files

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Needs: Needs Node.js

Security

No issues found by our static scan· No issues found

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Created by
Seth Hobson
Publisher
Seth Hobson (Community)
Source
wshobson/agents · plugins/llm-application-dev/skills/vector-index-tuning · commit 4bde316
Licence
MIT License (redistribution allowed)
  • LICENSE (added by AutoGineer from LICENSE)
  • ATTRIBUTION.md (written by AutoGineer)
Synced

AutoGineer is not affiliated with or endorsed by Seth Hobson or Seth Hobson.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download vector-index-tuning.zip