MLOps news and analysis on Web Pulse: 3338 curated articles tracking the latest MLOps developments, tools, and research, updated continuously from vetted sources.
AT&T now routes roughly 40% of its internal AI requests through open-weight models, cutting AI coding costs by 56% while absorbing only a 2% decline in output quality, the company said. The telecom giant processes approx…
A developer re-engineered an autonomous operational coordinator's processing pipeline to replace vector-database RAG with Hindsight's persistent agent memory, which extracts entities, temporal relationships and causal ch…
Local Minutes fine-tuned Qwen3.5-4B with LoRA on 24,000 examples generated by Claude Opus 5.5, completing one training pass on a single rented H100 GPU in 1.6 hours for about $7, after Gemini 2.5 Flash-Lite's retirement …
A team at HackwithHyderabad 3.0 built OnCallMemory, an AI incident-response agent that uses Vectorize Hindsight persistent memory to help on-call engineers reuse knowledge from previously resolved production incidents. T…
Researchers submitted TokenCast, a method that learns a composable cost representation for each execution segment of an LLM agent run, forecasting token consumption before and during execution without additional LLM call…
Google announced general availability of GKE Agent Sandbox optimized for reinforcement learning, along with an Agent Sandbox RL orchestration SDK and native integrations for RL gyms and harnesses, claiming 45x faster san…
A developer working on harness and charter engineering reports that giving a coding agent a single ticket with no charter produced a run that passed only 1 of 8 checks, while adding a four-lever charter — instructions, p…
A developer argues that instrumenting AI agents' chain-of-thought reasoning alters the behavior being measured, an observer effect that makes reasoning traces unreliable as evidence. The writeup distinguishes prompt-leve…
A developer outlines an architecture decision record for a unified image-generation API layer that places a narrow internal contract in front of model adapters, requiring bounded telemetry and evidence separation before …
A developer built `ivar graph`, a cross-repository code graph for coding agents that keeps a shared SQLite base index of each repo's default branch and per-feature layers holding only changed files, so agents working on …
Replaying 30 real coding-agent tasks on both a baseline model and a candidate model pins the quality gap to ±6 points at a cost of about $55, according to a Tuneloop analysis by Bharath Bhat. The method replays frozen ta…
A developer built WARROOM X, an incident-intelligence agent that uses Hindsight's persistent agent memory to store resolved incidents, root causes, resolutions and lessons, then recalls relevant history before reasoning …
An agent running Claude Code on Claude Opus 5 completed the same Postgres query-optimization task across three MCP servers — Supabase, Neon and Postgres MCP Pro — in times ranging from 143 seconds to 513 seconds, accordi…
A solo engineer running the live-trading system TopSet used AI coding agents including Claude Code, Cursor and Copilot to produce roughly 1,600 pull requests over 21 months, gating every merge on a green build of about 7…
A developer described splitting a headless AI agent server into separate ingress, Redis queue, and worker processes after discovering that containers reported as running while background agents silently hung, causing web…
A developer designed RECALL, a memory-driven incident-response agent that combines current alert data with historical incident knowledge to diagnose engineering failures. The workflow normalizes incoming alerts, retrieve…
A developer demonstrated wrapping Genkit flows inside senro pipeline steps, turning each Gemini model call into a graph node with its own log, state, retry policy and cache entry. The approach, tested against the real Ge…
CData Software Inc. introduced Connect AI Gateway in early access today, a platform that lets IT teams register models, MCP servers and agents and set policies governing their access to enterprise data. In a company-run …
A developer outlines a cost-attribution approach for AI coding assistants that tags token usage to specific features and tickets rather than teams or months, using Claude Code's locally written JSONL session transcripts …
Slack's engineering team published findings from over 200 automated agentic test runs comparing three architectures: an agent driving Playwright via Model Context Protocol, an agent issuing Playwright CLI commands, and L…