IDEAS, DECISIONS & LESSONS
Notes on AI, architecture and building useful products.
279 articles
20 Jun 2026 · 8 min read
agents
A practical guide to building an AI-powered monitoring dashboard that shows EC2 health, web platform status, CI/CD quality checks, and uses Claude to interpret anomalies in plain English.
Read more18 Jun 2026 · 7 min read
agents
After deploying multi-agent AI systems at scale, here are the failure patterns we hit and the resilience patterns that saved us. Practical guide for engineering teams.
Read more12 Jun 2026 · 6 min read
ai
Google just open-sourced the first credible text diffusion model from a major lab — DiffusionGemma-26B generates 1,000+ tokens/sec by abandoning autoregressive token generation entirely. Here's how Uniform State Diffusion actually works, when this architecture matters, and what it means for building with open-weights models.
Read more12 Jun 2026 · 5 min read
agents
MAF 1.0 merges AutoGen and Semantic Kernel into one production SDK — but the real story is CodeAct, which cuts token consumption 63.9% by having the model write Python programs instead of making individual tool calls. Here's what changes for .NET teams building agentic systems.
Read more11 Jun 2026 · 7 min read
agentic
Google's I/O 2026 announcements mark a shift from conversational AI to autonomous agentic systems. DiffusionGemma delivers 4x faster text generation. Gemma 4 broadens the open-weights ecosystem. And the new developer APIs signal where Google is pushing the industry next.
Read more11 Jun 2026 · 6 min read
ai
Anthropic just made its most powerful model tier publicly accessible. Claude Fable 5 and Mythos 5 represent a fundamental shift — not just in capability, but in who gets access to frontier AI. Here's what changed, what the safety trade-offs look like, and what it means for teams building on Claude.
Read more10 Jun 2026 · 5 min read
ai
Google's Gemini 3.5 Flash hit general availability with 4x faster inference than competing frontier models, a 1M token context window, and $1.50/$9 pricing. It outperforms Gemini 3.1 Pro on coding and agentic benchmarks. Here's the honest assessment of what this means.
Read more10 Jun 2026 · 5 min read
ai
Google's TurboQuant research reduces KV cache memory overhead by ~100x using a two-step algorithm combining PolarQuant vector rotation and Johnson-Lindenstrauss compression. This could make 2M-token context models economically feasible for many more teams.
Read more09 Jun 2026 · 4 min read
agents
Anthropic's Claude Opus 4.8 ships with Dynamic Workflows — a system where one Claude orchestrates up to 1,000 parallel subagents in a single session. The Bun-to-Rust port in 11 days is the proof of concept. Here's what this means for real engineering teams.
Read more
