Coding agents beat hand-built data agents by 37 points with 4x fewer turns, paper finds
CShorten30 · x · 2026-09-08
A paper by Liana Patel, Matei Zaharia, Ion Stoica and others asks what systems research remains for data agents under the Bitter Lesson.
- Tracking two years of data agent evolution, they find general-purpose coding agents beat hand-designed data agents by up to 37 points with 4x fewer turns.
- They argue system layers built to compensate for model limitations will increasingly be absorbed by the models themselves.
- Enduring opportunities lie in persistent semantic context: curated contextual information about the data environment served across many queries.
- Future data systems should natively serve persistent semantic context as a first-class abstraction, opening research into efficient context data structures, storage, compression, and semantic consistency protocols for integrity over huge knowledge corpora.
More from coding & agent
- Have your agent prototype in a throwaway branch before the real build — serious_mehta · 2026-09-08
- Founder builds Command & Conquer-style RTS dashboard to monitor AI agent sessions — davidhoang · 2026-09-08
- Warp CEO: we shipped an AI coding feature 6 months before Claude Code — holding back is my biggest regret — vikvang1 · 2026-09-08
- Utilify MCP lets AI agents compare and sign up for Texas utility plans by ZIP code — modelcontextprotocol · 2026-09-08
- Building a Kinesin ATP animation with Blender and Unreal MCPs — Mister-Fordo · 2026-09-08
- A developer proposes an informal agent-native mathlib, tentatively named mathgraph — Sauers_ · 2026-09-08