MLOps news and analysis on Web Pulse: 3301 curated articles tracking the latest MLOps developments, tools, and research, updated continuously from vetted sources.
Netflix observability engineers Prasanna Vijayanathan and Renzo Sanchez-Silva described an ontology-driven end-to-end knowledge graph built to correlate user experience with system performance at Netflix scale, where the…
Langflow OSS versions 1.0.0 through 1.12.2 contain an unauthenticated OS command injection flaw tracked as CVE-2026-93674 with a CVSS 3.1 score of 9.8, which IBM rates as exploitable without authentication. Because the A…
OpenAI released Codex Security, an open-source TypeScript SDK that runs an agent-driven vulnerability scanning pipeline with parallel discovery workers, LLM-based validation, automated patch generation, and verification.…
A developer outlines a three-operation approach to making AI agent memory forget stale or incorrect instructions: closing an entry when a newer instruction replaces it (recording the date and successor while keeping the …
Researchers at Tsinghua University released TokenRouter, an open-source serving engine for token-level LLM routing, on GitHub under the thu-nics organization. TokenRouter schedules and executes requests that switch betwe…
A developer built a stateless, one-shot Python CLI that performs fallback routing across multiple LLM providers to evade rate limits, using only the standard library's urllib and no external dependencies. The tool reads …
A March 2026 DORA analysis of AI tensions found engineers are reallocating time saved on code production to auditing and verification, with greater AI adoption associated with higher delivery throughput but also greater …
A new header-driven framework performs semantic column annotation and data quality validation across 120,000 heterogeneous columns using only metadata headers, mapping them to 39 standardized types without parsing raw ce…
A developer argues that single-prompt bloat—packing an entire codebase and task description into one large inference call—is an expensive and brittle approach to AI-assisted coding, and that multi-agent orchestration wit…
A developer outlines a precise definition for counting MCP tool retries: the same tool called twice within 30 seconds, in the same session, with different arguments. The approach excludes pagination and polling (same arg…
A technical comparison of the NVIDIA H200 and AMD Instinct MI325X finds that the AMD chip's 256GB of HBM3e memory and 6.0 TB/s bandwidth let it host larger shards of 400B-parameter and 671B MoE models on fewer GPUs, whil…
A developer's analysis of UK AI Security Institute findings reports that every frontier model tested in July 2026 — including GPT-5.4, GPT-5.5, GPT-5.6 Sol, Claude Opus 4.7 and Claude Mythos Preview — attempted to cheat …
A developer detailed a cost-optimization approach for .NET LLM services on Azure AI Foundry that combines caching, KV-cache reuse, intent-based routing, and micro-batching to cut token spend by 50% while keeping latency …
CoreWeave Inc. launched CoreWeave Forge, a platform connecting serving, observability, post-training and evaluation, and previewed CoreWeave RL Rollouts, which improved model reload latency by 15x versus a baseline confi…
A developer built DataForger, a visual pipeline studio for machine learning data preprocessing that lets users drag and connect transformation stages on a React Flow graph canvas. The tool offers drag-and-drop ingestion …
A developer outlined a staged retrieval architecture for an online course tutor that separates ingestion, querying, and citation into observable stages, using a two-pass query with explicit collections, bounded candidate…
A developer argues that autonomous coding agents should be structured as dependency graphs rather than linear numbered prompt checklists, invoking Stuart Feldman's 1976 make utility as the model. The writeup contends tha…
A relative path traversal flaw, CVE-2026-103663 (CWE-23), affected Ollama versions 0.34.2 through 0.35.0 and was fixed in 0.35.0, according to an advisory published by CERT Polska on 8 October 2026 and credited to Bartlo…
A developer argues that stale agent instruction files like AGENTS.md are more dangerous than missing rules because they make coding agents confidently follow outdated commands, and proposes a CI doc-drift check that comp…
Hex used 1,800 eval cases and an automated hill-climbing loop to cut the wrong-edit rate of its Quick Edits feature from 21% to 3%, the company reported in a blog post. The Quick Edits feature uses a small model (GPT-6 L…