Start here
How production agents handle stale approvals and uncertain writes
Connect authority, observability, containment, and ownership.
Read the articleAI in production
Understand AI capabilities, evaluate a bounded workflow and verify what its actions actually changed.
Write an execution contract
Recovering an interrupted agent skill
Describe the tool inputs, permitted actions, and failure handling.
Read the guideEvaluate the result
Testing an incident-response agent against the service it must recover
Build repeatable cases and an independent recovery gate.
Read the guideAccount for the cost
Where an AI task’s tokens, time and cost go
Include retries and context limits in the operational picture.
Read the articlePut it into practice
Related AI news
News ·
Anthropic takes its internal AI evaluations offline
Anthropic cut live internet access from internal evaluations after agents acted on real systems. The operational question is how to restore realism without restoring broad authority.
News ·
Haiku 5.5 lowers prices and changes what a log classifier must handle
Claude Haiku 5.5 brings lower token rates, adaptive thinking and new request constraints. A log-classification example separates integration failures from unsupported diagnoses.
News ·
What Nano Banana 2.1 changes in an image-generation service
Google's new image model rejects familiar generation settings. Follow a thumbnail job through its outgoing request, retries and delivery to evaluate the upgrade.
More on ai in production
Explainer · 5 min read
Scheduled AI work needs an execution contract
Bind the schedule to a model, tools, input state, destination and acceptance evidence, then reconcile the actual run before retrying or declaring success.
Explainer · 5 min read
Bit-exact decoding turns inference speedups into testable changes
Compare reference and speculative paths for token-level equivalence, then run a separate capacity trial so speed cannot conceal a serving-contract change.
Guide · 5 min read
Build an inference compatibility suite before changing the runtime
Pin the model artifact, template, runtime and hardware, then exercise fixed protocol tests and preserve raw differences before approving a serving change.
Explainer · 5 min read
Coding agent benchmarks: what actually won?
A coding agent's tools and testing policy help determine its benchmark result. Understand what was compared before choosing a setup for your team.