Engineering Articles
Production engineering practices: Open source contributions (DuckDB, DataFusion), testing strategies, and lessons from scaling data systems.
One Billion Rows
At 1 billion rows, every shortcut comes back to collect interest. Here's how we achieved sub-second queries with near-realtime ingestion.
Upside-Down Architecture
Most analytics queries scan <100MB. We explore a hybrid architecture where compute moves between servers and your local machine, powered by DuckDB and DuckLake.
Lessons from Search
Real-time analytics faces the small-file problem search engines solved. DuckLake's tiered compaction brings those merge strategies to streaming analytics.
From STDIO to OAuth
How MCP evolved from local stdio to OAuth 2.0 for cloud-scale AI, using Dynamic Client Registration for secure agent access.
NetQuack 4000x Faster
We rewrote NetQuack DuckDB extension, replacing regex with character parsing. Result: 4000x faster—37 seconds down to 0.012 seconds.
Upstreaming with AI
How we contributed 17 upstream PRs in 90 days—where AI accelerated our workflow, what we learned, and practical tips for open source success with AI assistance.
Under the Hood: Agents
Altertable agents think ahead. Powered by custom lakehouse and MCP, they monitor, investigate, and act on your data autonomously.
Relearning the Craft
AI-first development changed everything—prompts are code, green CI means nothing. Here's how we're adapting and relearning.
Behind the Stack
Our tech stack: Rails for backend velocity, Rust for high-performance ingestion, React and Urql for UIs, Iceberg for data lake, Trino for federated queries.
Building Trust, Byte by Byte
Trust is our foundation. Drawing on Algolia, Front, and Sorare experience, we build a data platform where security comes standard.
Interested in learning more about Altertable?
Apply for AccessStay Updated
Get the latest insights on data, AI, and modern infrastructure delivered to your inbox
Our Manifesto
The stack is broken: slow, fragmented, reactive. Our manifesto is a call to wake it up. It's time for data that works like your best teammate: always on, always thinking, always ready to help.
Read Manifesto