MLOps news and analysis on Web Pulse: 1306 curated articles tracking the latest MLOps developments, tools, and research, updated continuously from vetted sources.
A site reliability engineer at an unnamed company describes an operations architecture that uses an AI agent to read metrics, logs, and Kubernetes state and propose changes via pull requests, with no direct write access,…
XAI's terminal coding agent Grok Build now supports full observability via OpenTelemetry, emitting metrics and structured events that can be monitored in SigNoz. The integration requires setting the GROK_EXTERNAL_OTEL ma…
A developer has introduced ATMAR (Agentic Tracker Memory And Retrieval), a workflow framework that shifts continuity from chat sessions to project-root trackers to improve long-horizon AI-assisted software development. T…
A developer detailed how Langfuse provides the trace/span/generation hierarchy needed to debug multi-step agentic LLM workflows, contrasting it with observability tools that only capture top-level calls. The developer co…
ASOS, the online fashion retailer, is hiring a Platform Engineer for its Data Science & AI Platform in London, UK, with a hybrid work model. The role involves building and evolving infrastructure for data engineering and…
OpenCode, a terminal coding agent, stores session metadata in a local SQLite database, and a dlt pipeline moves that data into DuckDB for analytics. The pipeline, built by Alena Astrakhantseva and Alexey Grigorev in a Da…
A developer's field guide benchmarks DFlash, a block-diffusion speculative decoding method, on 30B-scale models, arguing that tokens per second can mislead. The guide highlights DFlash's architectural differences from EA…
Towards AI co-founder Louis-François Bouchard warns that coding agents can speed up software development but only if used with proper context, tests, reviews, and task boundaries, cautioning against shipping agent output…
Roy Libman's essay proposes a six-level autonomy scale for AI agents operating Kubernetes, modeled on the auto industry's self-driving levels. The framework uses the '3am test' to separate levels 3 and 4, asking whether …
A developer built a model catalog drift monitor for OpenAI-compatible Chinese AI APIs, designed to catch changes in model names, context windows, output limits, and pricing before they cause incidents. The monitor checks…
Decision engine observability connects technical telemetry to business decisions, enabling teams to trace which model ran, which version was deployed, and what output was returned. The approach uses versioned event schem…
An AI integration consultant detailed a week of work helping a B2B client add an LLM-based assistant to its stack, emphasizing that data plumbing, not the model, is the hard part. The consultant mapped data sources, chos…
A developer has released mcptoon, a 50KB CLI tool with zero dependencies, to address five common pain points developers face when using MCP servers with AI coding agents like Claude Code, Cursor, and Codex. The tool redu…
A developer reports that an unattended coding agent spent most of its budget on a frontier model for trivial tasks like docstring regeneration and version bumps, while a subtle bug received the same expensive treatment a…
A developer has detailed a risk-based routing workflow for AI coding tasks, arguing that the key question is not which model is best but which tasks require the strongest model. The approach assigns tasks to free-tier mo…
MonkeyCode's outreach describes a shadow CI lane that filters agent-generated patches before human review, checking that patches apply cleanly, pass a narrow test gate, and stay within declared file scope. The reference …
A developer who integrated a free AI coding model into a CI pipeline discovered that free endpoints can fail silently, returning empty completions with HTTP 200 and causing the pipeline to commit unusable changelog entri…
SkyPilot, an open-source framework, enables users to schedule jobs across multiple Slurm clusters, Kubernetes, and cloud providers from a single interface, addressing the pain points of managing multiple clusters manuall…
Dr. Samson Tanimawo, founder and CEO of Nova AI Ops, explains the concept of multi-agent SRE, where incident response is decomposed into specialized AI agents that coordinate via typed artifacts. He argues that this appr…
A developer reports a 55% reduction in test suite runtime from 96.2 seconds to 41.1 seconds by addressing three issues: redundant seeding, Xdebug overhead, and test impact analysis timeouts. The same developer also fixed…