Our point of view on shipping AI — page 4 of 10.
- Dev Tools5 min
Your GitHub Copilot Settings Change Meaning on October 22. Unconfigured Will Mean On.
By Petru Popa ·GitHub added a global default policy for Copilot Business and Enterprise features on September 24 and starts enforcing it on October 22. The changelog lists three choices without saying which one is preselected. The docs do: enabled. Every eligible feature you never configured, including MCP servers in Copilot, turns on unless you decide otherwise in the next four weeks.
Read - Dev Tools6 min
Node 20 EOL Reached GitHub Actions on September 23. Your node20 Actions Kept Passing.
By Petru Popa ·Node 20 went end of life on 30 April. GitHub Actions removed it on 23 September and killed the opt-out. Workflows calling node20 actions did not fail: the runner rewrites the declaration to Node 24 and leaves one warning at the end of the job. Over nine thousand public action.yml files still declare node20. The date on a deprecation notice is not the day your pipeline changes.
Read - AI Engineering6 min
To Run an LLM Locally in Transformers, the Metal Kernel Has to Be There
By Petru Popa ·Hugging Face shipped a path that runs llama.cpp quants inside transformers without dequantizing them. The packed path needs a Hub kernel whose only published driver family is metal, it forces float32, it covers Qwen3.5 and nothing else, and the fallback when the kernel is absent costs you the whole quantization.
Read - Dev Tools6 min
Copilot Custom Agents Pinned to Retiring Models: Aim at the LTS, Not the Suggested Alternative
By Petru Popa ·GitHub will retire GPT-5.5, GPT-5.4, GPT-5.4 mini, GPT-5 mini, Gemini 3.7 Flash and Grok 4.5 from every Copilot surface on 19 October. Over a thousand public custom agent files pin one of them. The docs table that tells you where to migrate still points four retired models at GPT-5.5, and the replacements it names have historically lasted months. The one target with a written commitment is the LTS model.
Read - AI Engineering6 min
DSpark Speculative Decoding: The 4x Was Measured on Math, and Tool Calls Accept About Half as Many Draft Tokens
By Petru Popa ·The vLLM team's DSpark drafter for Kimi K3 quadruples single-stream decode speed on math reasoning. The Hugging Face cards for the four published DSpark drafters show tool-calling prompts accept 43 to 59 percent as many extra tokens per step as math does, every published speed chart is a math chart, and vLLM's confidence-scheduled verification, the half of DSpark built for high concurrency, defaults to off.
Read - AI Engineering6 min
The Fastest Local LLM on a Mac Ships From a Repository Two Days Old
By Petru Popa ·Splash is an Apache-2.0 inference engine for Apple silicon that beats every engine Inco AI measured against it, on one machine, on two models. The repository behind it was created on 18 September and has twelve commits. Here is what the issue tracker and the model packages say about adopting it.
Read - Dev Tools6 min
Copilot's New Agent Observability Counts MCP Connections, Not Tool Calls
By Petru Popa ·GitHub added five breakdown arrays and five distinct-use counters to the Copilot usage metrics API on 17 September. The documentation says the MCP counter measures connection attempts rather than tool calls, and that anything your organization built itself is grouped under a single other row.
Read - Dev Tools6 min
GitHub Set a November 2 Deadline for pull_request_target. The Breaking Change Landed in July.
By Petru Popa ·Workflow execution protections reached general availability on September 17 with a default policy that blocks pull_request_target in public repositories from November 2. The change that actually breaks builds shipped to every moving actions/checkout tag back on July 20, and the two gates are independent of each other.
Read - AI Engineering5 min
Chord's INT4 MoE Kernel: Up to 2.15x per Layer, 4 to 8 Percent for a Self-Hosted LLM
By Petru Popa ·vLLM and Novita AI announced Chord, an INT4 mixture-of-experts kernel for Kimi K2.x, with per-layer speedups up to 2.15x. The end-to-end run in the same post shows 4.1 to 8.0 percent more decode throughput on eight H200s, the baseline is a Humming release from July, and the install gives vLLM's humming dependency a second owner.
Read - Dev Tools5 min
Copilot's New Cost Tiers Draw From a Pool That Spans 25x
By Petru Popa ·GitHub added three cost-and-quality tiers to Copilot auto model selection. The model table in the docs repository shows every tier drawing from one pool whose per-token prices span 25x, the discount for using auto at all is 10%, and the two controls that actually bound spend are set somewhere else entirely.
Read - AI Engineering6 min
WeKnora Is Enterprise RAG. Seven of Its Ten Security Advisories Hit the Agent.
By Petru Popa ·Tencent's WeKnora is an open-source enterprise RAG platform trending this week. Seven of its ten security advisories sit in agent tools and none in retrieval, while the quick-start .env file still ships fixed encryption keys with registration open.
Read - AI Engineering6 min
Self-Hosted AI Now Keeps Its KV Cache on Disk, and vLLM Never Expires It
By Petru Popa ·vLLM's tiered KV cache offloading writes prompt-derived blocks to a filesystem or an S3 bucket. The v0.29.0 tier code has no age or capacity expiry, creates block files mode 0644, and names them with a hash any instance can recompute.
Read