Recall != consumption. Same anti-pattern reproduced across sessions despite recall hitting the right files. Three layers of fix and a capability-driven governance plan that scales without templates.
Compass v1.1.0 · the recall consumption fix
We shipped nautilus-compass v1.1.0
12 hours after v1.0.0. v1.0.0 was the public stable cut. v1.1.0 fixes a
class of failure that v1.0.0 surfaces but does not catch · which we
caught in our own usage 5 hours after launch.
Compose Multiplatform 1.11.0 Is Now Available | The Kotlin Blog
shopify.engineering·Autoresearch isn’t just for training models (2026) - Shopify
Exclusive: a conversation with Tibo from Codex on what your company has to become when the model can actually do the work
Introducing Composer 2.5 · Cursor
Software Forgets: Agent Traces Are the Memory
Cursor's Composer 2.5 matches Opus 4.7 and GPT-5.5 benchmarks at a fraction of the cost
Pushing the boundaries of what's possible with FlutterFlow
OpenAI made Codex smart enough that the bottleneck moved. Most people haven't noticed where it went.
Two weeks of dogfooding Engram, Weaviate's memory product, in daily Claude Code sessions. This surfaced where a dedicated memory…
: This CVSS 10.0 RCE vuln has been patched, automatically for some, so better check those workflows
How we measure crash and OOM rates, debug memory issues with top-down and bottom-up strategies, and ship mitigations and…
Claude Sonnet 4.6, Gemini 3.1 Pro, Lyria 3, Qwen 3.5 397B-A17B, Grok 4.20, Google's Photshoot in Pomelli, Claude security review,…
Most production AI features are still glueing prompts together with string concatenation in some random Node service. The prompt…
Claude Opus 4.7, Claude Design, Codex everything app, GPT-Rosalind, Qwen 3.6-35B-A3B, Gemini 3.1 Flash TTS, Lyra 2.0, Happy…
