TD

Thierry
Damiba

I build retrieval and memory systems for agents: hybrid search, eval harnesses, and production demos.

Featured work

All projects →
Observability
Agentglass (Switchyard)

Chrome DevTools for AI agents: trace viewer, attribution overlay, diff mode, critical path analysis, portable repro bundles.

Eval + Tuning
MaxQ

Retrieval evaluation + tuning workflow: ingest, index variants, eval suite, recommendations. Built for repeatability.

Writing

All posts →
Benchmarks
Building Better LLM Benchmarks with Prime Intellect

NBA stats, Balatro, CAMB, and the cost of curiosity.

Agents
How to Scale Agent Retrieval Without Guessing

From Q&A chatbots to systems that plan, retrieve, act, and verify.

Retrieval
Balancing Relevance and Diversity with MMR Search

How native MMR breaks the echo chamber of similar results.

Talks

Google Cloud Next
Data on Kubernetes: Qdrant + GKE

Persistent storage, caching strategies, checkpointing. Led to GKE Data Cache feature.

Haystack US 2025
MiniCoil: A Hybrid Sparse Retrieval Model

Sparse retrieval architecture for production hybrid search.

AWS AI Loft, SF
Best Embedding Model for Your RAG App

199 attendees. Evaluation methodology for embedding model selection.

ODSC AI West
Powering AI with Vector Search

Hands-on training session.

LlamaIndex + Qdrant
Beyond the Tutorial: Agents with LlamaIndex & Qdrant

RAG-enabled agents for complex queries across multiple data modalities.

Deutsche Telekom
Multi-Agent Platform: LMOS

Platform powering 2M+ conversations across three countries.

Production RAG
A Data Driven Approach to Productionizing RAG

Practical strategies for prototype to production.

Podcast
AI Ketchup: Vector Databases and Agentic Systems

AI security, vector databases, and agent reliability.

Video

About

I work on retrieval, evaluation, and agent reliability at Qdrant. I help teams ship production-grade semantic search, hybrid retrieval (dense + sparse), and agentic RAG pipelines with clear latency, cost, and quality tradeoffs.

Previously: ML systems for the Department of Homeland Security at SAIC (classification + forecasting). English and French. University of Maryland, College Park.

Benchmarking and eval harnesses
Hybrid retrieval architectures
Agent observability and attribution
Applied prototyping to docs, talks, OSS
Google Cloud Next, Haystack, ODSC, AWS
San Francisco. Baltimore originally.

Contact

Fastest on X. Open to talks, consulting, and collaborations. San Francisco, PT.