blog category
Artificial Intelligence
Technical analysis of AI agents, local inference, evaluation, and the systems required to deploy them safely.
- post
Agent Skills Joins AAIF: What Engineering Teams Should Change
Treat portable agent instructions as reviewed dependencies, with runtime permissions and evaluations kept explicit.
- post
Anthropic’s Agent Incidents: Put Permissions Outside the Model
What the October 9 disclosure means for tool permissions, browser automation, and agent evaluation design.
- post
Beyond Autocomplete: Architecting Web Applications with Claude and Codex
Why inline code generation breeds technical debt—and how senior web developers orchestrate Claude and Codex for robust system boundaries and state machines.
- post
Claude Opus 5.5: Why the Model Recommendation Just Changed
Opus 5.5 now performs at Claude Fable 5.1's level on most work while costing 40% less than the old Opus 5, and per-token it's now cheaper than Opus 5 was three weeks ago. That changes the advice from this site's last model comparison.
- post
Claude Fable 5.1 vs Opus 5 vs Sonnet 5: Which Model Should You Actually Use?
Fable 5.1 costs twice what Opus 5 does per token and shows only a 3-point gain over Fable 5 on CursorBench, but a 33-point jump on agentic coding benchmarks. The right pick depends entirely on which of those two numbers describes your workload.
- post
Google's Open Knowledge Format Doesn't Replace Vector Search — It Fixes What Embeddings Were Never Good At
OKF v0.2's actual spec, read against the RAG failure modes it's being sold as a cure for — and the hybrid architecture that follows once you separate the two problems.
- post
Architecting Production Pipelines for Autonomous AI Coding Agents: Lessons from the Engineering Frontline
A practical software architecture guide to context window assembly, sandboxed runtime environments, automated lint/build feedback loops, and human-in-the-loop governance.
- post
Running Small Language Models (SLMs) on Edge & Browser Runtimes: WebAssembly, WebGPU, and Local Inference
How client-side AI execution using WebGPU and WebAssembly is eliminating cloud API latency, guaranteeing zero data egress, and transforming browser developer tools.