October 28, 2026 · AI Engineering Reflection
The prompt cache is an architecture decision
Why repeated AI workflow context needs an owned, versioned boundary before caching can improve real product economics.
Read post
September 28, 2026 · AI Engineering Reflection
The trace should travel with the change
Why AI workflow changes need compact trace and evaluation evidence that can be reviewed before release and compared in production.
Read post
September 21, 2026 · AI Engineering Reflection
The first wrong tool call deserves its own metric
Why agent teams should isolate the first causal tool or response failure instead of debugging every downstream symptom as a separate defect.
Read post
September 14, 2026 · AI Engineering Reflection
An agent session is not a workflow
Why durable agent sessions only become dependable products when state, recovery, retries, and side effects are designed as an inspectable workflow.
Read post
September 10, 2026 · AI Engineering Reflection
An AI approval is not a merge policy
Why AI pull-request approval should begin as a measured review signal, while teams keep branch protection tied to explicit evidence and human accountability.
Read post
August 31, 2026 · AI Engineering Reflection
Agent access should be earned, not assumed
Why agent capability should expand through evidence, bounded environments, and explicit approval instead of arriving as a default permission set.
Read post
August 24, 2026 · AI Engineering Reflection
A passing test is not an AI quality strategy
Why an AI-generated green build or coverage increase is only useful when it comes with evidence of behavior, review value, and quality movement.
Read post
August 17, 2026 · AI Engineering Reflection
A plugin is an integration contract
Why portable skills and MCP configuration matter only when they become versioned, governed workflow contracts.
Read post
August 11, 2026 · AI Rollout Strategy
Your AI rollout needs tiers, not one default assistant
Why this week's clearest signal was role-based AI access, budget tiers, tool allowlists, and review depth becoming policy surfaces.
Read post
August 3, 2026 · AI Engineering Reflection
The second build is where AI products get real
Why this week's strongest signal was that production AI work is shifting into telemetry, policy, evals, and operating structure.
Read post
July 27, 2026 · AI Product Delivery
One reviewable path beats a very smart agent
Why last week's tooling signal pushes me toward narrower, inspectable AI workflows instead of broader autonomy.
Read post
July 20, 2026 · AI Delivery Strategy
The repository is becoming the AI rollout unit
Why enterprise AI adoption is shifting from license counts toward repo-level readiness, controls, and evidence.
Read post
July 13, 2026 · AI Engineering Practice
Longer agent runs need shorter feedback loops
Why stronger AI agents increase the value of tighter eval, approval, and review loops instead of removing them.
Read post
July 6, 2026 · AI Product Architecture
The AI workbench is becoming the product
Why this week's bigger signal was domain-specific AI systems packaging models, context, artifacts, and compute into one operating surface.
Read post
June 29, 2026 · AI Collaboration Systems
The next AI bottleneck is the team thread
Why the stronger signal this week was agents moving into shared queues, tickets, and channels rather than staying inside private chats.
Read post
June 22, 2026 路 AI Runtime Design
AI agents need commit boundaries
Why the deeper signal this week was the move toward staged effects, failure lineage, and cost accountability.
Read post
June 21, 2026 · AI Engineering Practice
Your agent needs operating documents
Why machine-readable instructions, capability catalogs, eval traces, and usage telemetry are becoming part of the real AI system.
Read post
June 15, 2026 · AI Platform Strategy
A model name is not a stable architecture
Why upstream changes in model access, permissions, and spend controls now shape system behavior as much as model quality.
Read post
