X

@cursor_ai

Rollouts: write a monitoring plan, watch deploys, catch regressions before users see them. Teams/Enterprise; trial credits for 10 days. post (https://x.com/cursor_ai/status/2102861817160904808) · 2,251❤ · 119 RT

7% token cost cut: same agent quality via tighter prompts, selective tools, caching, compressed reads. post (https://x.com/cursor_ai/status/2102786814633464159) · 4,125❤ · 168 RT

@poteto

Grok Bot game studio: started a studio on the galaxy livestream; recap, past streams, and first game linked. post (https://x.com/poteto/status/2102835770784694385) · 614❤ · 33 RT

@rauchg

Agent triad: brain (model/harness), hands (tools/computer/browser), files (memories/skills/repos); easy path is one stateful computer. post (https://x.com/rauchg/status/2102820148629614685) · 1,675❤ · 98 RT

ShellPerfBench: neurotic about shell startup; Opus 5.5 found .zshrc optimizations other models missed. post (https://x.com/rauchg/status/2102947745132924993) · 141❤ · 3 RT

@simonw

Datasette voice chat: voice conversation with a Datasette backup of his blog via datasette-mcp in the ChatGPT iPhone app. post (https://x.com/simonw/status/2102916558712705427) · 102❤ · 8 RT

Gemini 3.8 TTS: cheap multi-voice (2,000+ or clone your own); built a UI plus a pelican debate script. post (https://x.com/simonw/status/2102861892549279922) · 550❤ · 20 RT

@hwchase17

Proactive agents: run in the background; schedules are one way to get there (LangChain Managed Deep Agents cron). post (https://x.com/hwchase17/status/2102802015789142478) · 49❤ · 5 RT

@HamelHusain

AI Evals guide: public materials organized by problem type (humans pick a row; agents point). post (https://x.com/HamelHusain/status/2102807203967365549) · 789❤ · 93 RT

What are AI evals: tests that tell you whether an AI system is doing what you want; model benchmarks ≠ product evals. post (https://x.com/HamelHusain/status/2102805207289593987) · 79❤ · 10 RT

Bot reply slop: bots replying to each other are a firehose of slop. post (https://x.com/HamelHusain/status/2102757318320648242) · 33❤ · 0 RT

Silent

@martinfowler, @karpathy, @swyx: no original posts in window

Gaps

@xAI: API not authorized to read this account's posts (skipped)

GitHub trending

anthropics/financial-services (https://github.com/anthropics/financial-services): reference agents, skills, and data connectors for financial-services workflows (Python, +664★ today)
google/ax (https://github.com/google/ax): Google's open agentic orchestration runtime (Go, +1,543★ today)
davila7/claude-code-templates (https://github.com/davila7/claude-code-templates): CLI for configuring and monitoring Claude Code (Python, +389★ today)
BuilderIO/agent-native (https://github.com/BuilderIO/agent-native): framework for building agentic apps (TypeScript, +87★ today)
obra/superpowers (https://github.com/obra/superpowers): agentic skills framework and software development methodology (Shell, +474★ today)