MLOps news and analysis on Web Pulse: 3307 curated articles tracking the latest MLOps developments, tools, and research, updated continuously from vetted sources.
A developer has released Chloe, an open-source MIT-licensed TypeScript runtime for AI agents that lets developers set the level of autonomy for each step of a workflow, from plain API calls to model judgment, autonomous …
A six-level maturity model for running LLM systems in production places "confidence" at Level 3, arguing that an agent's most important output is a calibrated certainty score rather than its answer. The model composes th…
A six-level maturity model for running LLM systems in production begins with a "determinism layer" that contains the model to a single node so the surrounding system is ordinary, testable code, according to a technical p…
A developer built OpsPilot AI, a retrieval-augmented generation system that searches tenant-owned operational knowledge, answers with citations, and prepares GitLab issue proposals for human approval without granting the…
A developer released REA (Reverse Engineer Anything), an open-source Model Context Protocol server that orchestrates reverse-engineering tools such as Ghidra, Hopper, and dynamic tracers behind a single stateful API for …
A developer building RAG features in Python describes a staged retrieval architecture for travel itinerary caching that separates document and query-result caches, uses explicit collections (places, transport, lodging, s…
A GenePattern Copilot demo hallucinated a nonexistent module called scPerturb in front of a dozen medical students at a San Diego workshop, and the assistant's fully retrieval-augmented RAG pipeline subsequently scored w…
A benchmark of vLLM 0.30.0, SGLang 0.5.20, and TensorRT-LLM 1.2.1 on a single H100 SXM GPU with 8 CPU cores found no host-bound regime across Qwen 3 0.6B-32B models at batch sizes from 1 to 256, with CUDA graphs enabled,…
A ForgeWorkflows engineer argues that enterprises have built a structural blind spot by separating cloud infrastructure teams from AI/ML operations, leaving no single person able to correlate autonomous agent decisions w…
A developer outlines an architectural approach for debugging non-deterministic failures in AI-augmented state management, arguing that traditional replay-based debugging breaks down when LLMs sit in the state machine. Th…
LastPing, a monitoring service for cron jobs, CI pipelines and AI agents, rewrote all 50 of its MCP tool descriptions to comply with Anthropic's Claude Connectors Directory policy, after an audit found 34 of the 50 conta…
A developer behind Spread Out, a tool that turns a single URL into a week of scheduled social videos, carousels and podcasts, detailed the engineering lessons from two years of building it ahead of the V2 launch on Produ…
GrowthBook published a case study reporting that Fyxer, a UK AI email and meeting assistant company, ran 541 experiments in 12 months, with its four-person growth engineering team responsible for 360 of them, after conne…
NVIDIA's DSX Air node-based digital twin lets AI factory operators validate supported infrastructure configurations and software changes before physical hardware is racked, according to the company's description of the p…
AWS published a reference architecture that pairs AWS DevOps Agent with AWS Lambda Durable Functions, Amazon EventBridge, and Amazon Bedrock to automate incident remediation on AWS. In the workflow, EventBridge triggers …
Trigora launched a durable execution platform for long-lived AI agents and programs that recovers from committed continuation state instead of replaying completed execution history, using a technique it calls Transparent…
Lumen's Black Lotus Labs reported that PoeLLM, an ELF malware file named libgcrypt, has infected more than 2,100 internet-exposed servers running AI and developer tools including LiteLLM, Ollama, Gotenberg and Gitea, usi…
Postman launched Astropods, an agent operating system layer within the Postman AI stack that lets developers build, deploy, and monitor AI agents in production, the company said in a guide to the product. Astropods organ…
A developer outlines a load-testing methodology for applications built on third-party LLM APIs, arguing that conventional load-testing assumptions break down because external AI providers impose rate limits, bill per tok…
A developer built Origin, a repository interface that tracks AI-written code by agent, model, prompt, cost and review status, reporting that 93.9% of code in their own repo was written by Claude. Origin also stores per-r…