Our point of view on shipping AI — page 3 of 7.
- AI Engineering6 min
MXFP4 Quantization Did Not Improve the Model. A Second Teacher Did.
By Petru Popa ·A Hugging Face write-up reports a 4-bit MXFP4 model beating its own full-precision original on 7 of 9 benchmarks. The paper underneath it says the teacher was already MXFP4, that every number is a single run, and that the baseline which would settle the claim was never run.
Read - Dev Tools6 min
Log Retention Falls to 90 Days, and Two More Defaults You Did Not Choose
By Petru Popa ·From 1 October, GitHub applies artifact-and-log retention to checks, workflow runs, and statuses, which had been kept past 400 days regardless of the setting. Public repositories cap at 90 days with no value above it. Two other defaults moved in the same window, in opposite directions.
Read - Dev Tools6 min
Gang Scheduling Is Beta, and Beta Means Off
By Petru Popa ·Kubernetes v1.37 graduates gang scheduling to Beta through the Workload and PodGroup APIs. The feature gate that turns it on ships defaulted to false, and the DRA piece that makes it useful for GPU jobs is a second gate, also off, that depends on the first.
Read - Dev Tools6 min
Stacked Diffs Split the Change, Not the Reviewers
By Petru Popa ·maiao v1.4.0 brings Gerrit-style stacked diffs to six git hosts, weeks after GitHub put native stacked pull requests into public preview. The repository ships itself as a stack, which turns its own release into a measurement of what stacking buys.
Read - AI Engineering6 min
Agentic Memory Helps Until You Cap the Retrieval
By Petru Popa ·IBM Research reports gains of up to 16.1 points from feeding an agent guidelines mined from its own past runs. The technical report underneath contains a configuration of the same system that scores below the no-memory baseline, and the library that ships defaults to a retrieval backend no reported configuration used.
Read - AI Strategy6 min
To Block AI Bots, Longest Match Beats File Order
By Petru Popa ·Cloudflare's new Bot Preference Sync generates your robots.txt from the AI bot policy in your dashboard and prepends it to whatever your origin already serves. Prepending is not overriding, and the standard explains why a narrower Allow you wrote years ago still wins.
Read - AI Engineering6 min
Word Error Rate Fell 3.5 Points When Someone Fixed the Transcript
By Petru Popa ·Artificial Analysis hand-corrected 628 VoxPopuli clips and average model word error rate dropped 3.5 points with no retraining. A new set of probes explains the other half of that result: the models scoring best had learned to reproduce the errors.
Read - Dev Tools5 min
Write Access Decides Who Starts the Agent, Not Who It Hears
By Petru Popa ·GitHub shipped shared Copilot cloud agent sessions into Slack and Microsoft Teams in public preview. Only people with repository write access can trigger the agent, and guests cannot start or steer a session at all. But the docs also say the agent reads every message in the conversation. Those are two different boundaries, and only one of them is enforced.
Read - Dev Tools6 min
The OAuth Consent Screen Can Now Hand Back Less Than You Asked For
By Petru Popa ·Cloudflare added an optional_scopes field to third-party OAuth clients, so a user can clear individual permissions at the consent screen and the issued token carries only what survived. The announcement tells you to check the granted scope set. It does not say which field carries it, and neither does the client documentation. RFC 6749 does.
Read - Dev Tools5 min
The Mojo License Is Apache. The MAX SDK License Is Not.
By Petru Popa ·Modular put the Mojo compiler on GitHub under Apache 2.0 with LLVM exceptions. The same repository carries the Modular MAX Community License, whose terms are not the Apache terms and which never names the packages it governs. Nine merged pull requests across all of 2026 answers the other question a technical lead has.
Read - AI Engineering5 min
One NDCG Point for Twelve Times the Index
By Petru Popa ·Late interaction retrieval lands in Sentence Transformers 6.0 with a controlled benchmark behind it: same backbone, same data, one vector per token instead of one per document. The margin is about one NDCG point. The release notes show what that point costs in storage, and the model table shows which knob actually moves it.
Read - Dev Tools6 min
The Reviewable Unit Is the Pull Request, Not the Commit
By Petru Popa ·Wiz published research on an Actions script injection that leaked a Jira API token from a Snowflake runner. The pull request that introduced it carries a commit attributed to Copilot Autofix, and the diff removed the exact mitigation GitHub's own hardening page prescribes. The artifact cannot tell you which lines the model wrote — and that gap, not the model's error rate, is what your review process has to absorb.
Read