PPhying中文
Curated security research, vulnerabilities, advisories and tools for practitioners.

How we built ten custom subagents to tame a 500K-line Clojure codebase

metabase.com | blog | #claude-code | #metabase | #ai-coding | #clojure | #subagents | #context-management
📄 Raw .md⬇️ Download .md🔗 Original ↗

Summary

How Metabase built ten domain-expert subagents — markdown files of roughly 2,000-3,000 tokens each — to stop Claude Code paying a context tax across its 500K-line Clojure backend.
Published
Collected

Related coverage

blog · jfrog.com
Our AI Agent Now Has a Security Conscience: Introducing the JFrog Plugin for Claude Code
JFrog's plugin for Claude Code brings security scanning, package safety checks and governed MCP server management into the IDE, vetting what AI agents build and pull at development time.
blog · metabase.com
'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse
We asked 73 data people about AI Analytics: 59% are already using AI with their data, 44% want it all self-hosted, and 31% of poll voters don’t know what a semantic layer is. A small, unscientific snapshot from AI Analytics Week.
blog · metabase.com
We don't use PHP
We don't use PHP at Metabase, and we removed eMudhra-related certificates in a recent update.
blog · metabase.com
As humans write less code: workloads shift from human to agent verification with iterative test suites
How iterative test suites, human intuition, and unified data lineage can help verify agent-generated analytics.
blog · metabase.com
Making pivot tables faster with GROUPING SETS
Pivot tables used to require many queries and a lot of glue code to keep the totals consistent. Using SQL GROUPING SETS, we are now able to query the database for the whole pivot table at once.
blog · metabase.com
The bottleneck moved: how to build a semantic layer for AI Analytics
AI removed the querying bottleneck and created a verification one. Here's the semantic layer that makes AI-generated answers traceable, and how to build it.
Back