David Myriel
David Myriel

David Myriel

Notes on vector search, RAG and agent memory from building retrieval systems for companies. Previously DevRel at Qdrant; writing for LanceDB, Cognee and Tigris.

2026

Is RAG Dead? Long Context Windows versus Retrieval ↗
Qdrant
Aug
The Most Expensive ClickHouse Query Is the Restore ↗
Tigris Data
Jul
Where Does the Agent Live? ↗
Tigris Data
Jul
Every Tenant Has a Past: Evaluating LangGraph Agents ↗
Tigris Data
Jun
Introducing Soft Delete for Tigris Buckets and Objects ↗
Tigris Data
Jun
Cut Cognee's Vector Memory by 8x with Qdrant's TurboQuant ↗
Cognee
Jun
Own Your AI Context with Basic Memory ↗
Tigris Data
May
Build a Self-Updating Knowledge Base for Under $10 ↗
Tigris Data
May
The Immutable Agent ↗
Tigris Data
Apr
Introducing Agent Kit: Storage Workflows for AI Agents ↗
Tigris Data
Apr
Fifty Agents for the Price of One Bucket ↗
Tigris Data
Apr
Tigris Agent Plugins: Your AI Coding Agent Now Speaks Storage ↗
Tigris Data
Apr
Cognee's CLI Replaces MCP OAuth in 100 Lines ↗
Cognee
Apr
Agents Don't Need Another Protocol. They Need a Good CLI. ↗
Cognee
Apr
Introducing Multi-Region and Dual-Region Buckets ↗
Tigris Data
Mar
Event-Driven Multimodal Ingestion with Tigris and Mixpeek ↗
Tigris Data
Mar
Grounding AI Memory: How Cognee Uses Ontologies to Build Structured Knowledge ↗
Cognee
Mar
Test RAG Chunking Strategies With Tigris Forked Buckets ↗
Tigris Data
Jan

2025

Why Agent Memory Breaks (and How to Fix It) ↗
Cognee
Nov
How Cognee Builds AI Memory Layers with LanceDB ↗
LanceDB
Sep
LanceDB's RaBitQ Quantization for Blazing Fast Vector Search ↗
LanceDB
Sep
Building Semantic Video Recommendations with TwelveLabs and LanceDB ↗
LanceDB
Sep
Netflix's Media Data Lake and the Rise of the Multimodal Lakehouse ↗
LanceDB
Aug
LanceDB WikiSearch: Native Full-Text Search on 41M Wikipedia Docs ↗
LanceDB
Aug
What is the LanceDB Multimodal Lakehouse? ↗
LanceDB
Jun
Beyond Multimodal Vectors: Hotel Search with Superlinked and Qdrant ↗
Qdrant
Apr
Qdrant 1.14: Reranking Support & Extensive Resource Optimizations ↗
Qdrant
Apr
Qdrant 1.13: GPU Indexing, Strict Mode & New Storage Engine ↗
Qdrant
Jan

2024

Building a Facial Recognition System with Qdrant ↗
Qdrant
Dec
How Vector Databases Work and When They're Used ↗
ADMIN magazine
Dec
Best Practices in RAG Evaluation: A Comprehensive Guide ↗
Qdrant
Nov
Qdrant 1.12: Distance Matrix, Facet Counting & On-Disk Indexing ↗
Qdrant
Oct
Data-Driven RAG Evaluation: Testing Qdrant Apps with Relari AI ↗
Qdrant
Sep
Why vector databases aren't just databases ↗
InfoWorld
Sep
Qdrant 1.11: The Vector Stronghold ↗
Qdrant
Aug
Qdrant 1.10: Universal Query, Built-in IDF & ColBERT Support ↗
Qdrant
Jul
Blog-Reading Chatbot with GPT-4o ↗
Qdrant
May
Integrating Qdrant and LangChain for Advanced Vector Similarity Search ↗
Qdrant
Mar
Qdrant 1.8: Major Performance Enhancements ↗
Qdrant
Mar
RAG Is Dead. Long Live RAG! ↗
Qdrant
Feb
Best Practices for Massive-Scale Deployments: Multitenancy and Custom Sharding ↗
Qdrant
Feb
Need retrieval or agents built? I take a few consulting clients at a time. Message me on LinkedIn →

Projects

TopK product map
Every part of TopK on one page.
Multi-vector search tour
Interactive walkthrough of multi-vector retrieval.
Regex search tour
How fast regex search works with sparse n-grams.