Samuel Oakley

Systems engineer obsessed with vector search indexing, PostgreSQL connection pooling, and low-latency database replication. Writing about real-world database war stories.

Austin, TX Joined on Oct 17, 2026

Work

Database & Storage Engineer

Samuel Oakley

Oct 27

Cutting RAG Token Bloat: Compression Before Retrieval in Production Pipelines

#database #ai #python #backend

4 min read

Samuel Oakley

Oct 22

Keeping Vector Retrieval Stable While Your Knowledge Base Is Still Loading

#database #ai #python #backend

5 min read

Samuel Oakley

Oct 20

Benchmarking mem0ai/mem0: Memory Overhead and Recall Latency Under Concurrent Agent Workloads

#database #ai #python #backend

3 min read

Samuel Oakley

Oct 19

Integrating ByteDance deer-flow: Vector Memory Persistence Patterns for Long-Horizon Agents

#database #ai #python #backend

3 min read

Samuel Oakley

Oct 19

Production Vector Memory Integration with AutoGPT: HNSW Index Bottlenecks and Hybrid Search Trade-offs

#database #ai #python #backend

5 min read

Samuel Oakley

Oct 18

The Silent Recall Drop: pgvector HNSW Under Concurrent Writes

#database #ai #python #backend

3 min read

Samuel Oakley

Oct 18

Benchmarking Vector Search Latency: pgvector HNSW vs IVFFlat Under Concurrent Load

#database #ai #python #backend

4 min read

Samuel Oakley

Oct 17

Vector Search Latency Crisis: What I Learned Integrating RAGFlow's Memory Layer

#database #ai #python #backend

3 min read

DEV Community

Samuel Oakley

Badges

Writing Debut

Skills/Languages

Cutting RAG Token Bloat: Compression Before Retrieval in Production Pipelines

Keeping Vector Retrieval Stable While Your Knowledge Base Is Still Loading

Benchmarking mem0ai/mem0: Memory Overhead and Recall Latency Under Concurrent Agent Workloads

Integrating ByteDance deer-flow: Vector Memory Persistence Patterns for Long-Horizon Agents

Production Vector Memory Integration with AutoGPT: HNSW Index Bottlenecks and Hybrid Search Trade-offs

The Silent Recall Drop: pgvector HNSW Under Concurrent Writes

Benchmarking Vector Search Latency: pgvector HNSW vs IVFFlat Under Concurrent Load

Vector Search Latency Crisis: What I Learned Integrating RAGFlow's Memory Layer