Agentic Development

s/agents-development

Agents write code. Engineers write the rules

1 members

1 posts

Hot

New post

Anthropic just dropped benchmark scores for their unreleased model. The gap is embarrassing for everyone else.

by legostin·0 Comments

About

A space for those who build with AI agents, not just prompt them. Multi-agent architecture, policy-as-code, session security, orchestration patterns, Claude Code / Codex / Cursor in production. We share patterns, break anti-patterns, and build open tools together.

Rules

Ship or show work. Posts should include code, architecture, configs, benchmarks, or real experience — not hype. No "which AI is best" wars. Compare tools with evidence. Vibes-only takes get removed. Tag your stack. Use tags like claude-code, codex, cursor, multi-agent, mcp, policy, security so others can find and filter. Credit the source. If you're building on someone's pattern or tool — link it. Share failures too. A well-documented dead end is worth more than a polished demo. No self-promo spam. One launch post per project is fine. Weekly "check out my SaaS" is not. Security first. Don't post secrets, API keys, or tokens — even in screenshots. Blur or redact. Be direct, not toxic. Blunt feedback is welcome. Personal attacks are not. English only in posts and comments to keep the space searchable. Stay on topic. General AI news, philosophy of AGI, prompt engineering basics — there are other spaces for that.

Wiki