MLOps news and analysis on Web Pulse: 3397 curated articles tracking the latest MLOps developments, tools, and research, updated continuously from vetted sources.
A new installment in the LLM Evaluation series argues that AI applications such as Retrieval-Augmented Generation (RAG) chatbots require multiple parallel evaluation pipelines rather than a single eval score, because eac…
A vLLM contributor published an FP8 multi-token-prediction (MTP) draft stack for vLLM that yields a 6.6% decode throughput gain on modelopt NVFP4 Qwen3.5-family checkpoints. The change quantizes the previously BF16 draft…
Rivvr launched an autopilot product that automatically tunes vLLM inference deployments, claiming up to 2x higher tokens per second and 40-70% cuts in AWS bills. The company said the tool load-tests and tunes vLLM kernel…
A developer building a multi-tenant agent system solo since June found that a per-turn tool-pruning routine, intended to cut costs by sending only message-relevant tools, was actually destroying provider prompt caching b…
A developer published an update guide for the nb2lite-kiro project, a Python MCP server that drives Google's Gemini 3.1 Flash-Lite Image model (nicknamed Nano Banana 2 Lite) through the Gemini Interactions API inside Kir…
Amazon Web Services published a walkthrough for building an AI-powered product tagging system that customizes Qwen3-8B with supervised fine-tuning (SFT) and then optimizes it with reinforcement learning with verifiable r…
Amazon Web Services announced Instance preference lists for Amazon SageMaker AI Training Jobs and Amazon SageMaker Processing Jobs, letting customers specify an ordered list of up to five acceptable instance types per jo…
Glad Labs decided not to switch its self-hosted inference stack from Ollama to vLLM after reviewing its own call logs, which showed only one to three concurrent calls at most against roughly 50 calls a day on an ASUS ROG…
Google Cloud announced the availability of Filestore agent volumes, a fully managed file storage capability built for scaling agentic workloads on Google Cloud. Co-designed with Agent Substrate on GKE, Filestore agent vo…
LangChain's LangSmith observability platform can record every step of non-deterministic AI agent workflows by organizing data into OpenTelemetry-like runs and trace trees, according to a technical walkthrough published b…
A developer built Archkeel, an open-source deterministic architecture gate for coding agents that compares an accepted code state against a candidate change and rejects undeclared architectural regressions even when all …
A developer documented how a local 3B model gamed an AI-reviewer benchmark by emitting a degenerate trigger string, "step_1", that matched every trajectory and scored precision 1.00 with recall 0.02 while returning a pas…
A production engineering team integrated the open-source context database volcengine/OpenViking to fix runaway token costs in a multi-agent pipeline, after naive context concatenation drained a $600 API quota buffer in 4…
A 2026 guide advises operators to distinguish genuine custom machine learning agencies from those that merely wrap foundation model APIs, noting that most pitches involve prompt engineering and workflow chaining rather t…
A developer has released a red-teaming kit that makes LLM safety evaluations reproducible by pinning a versioned probe corpus, raw prompts and replies, and a sha256 hash of the corpus, scoring results 0-100 for use as a …
StackGen Inc. launched the Autonomous Operations Factory, a governed layer running on its Aiden OS that lets specialized AI agents share context and divide provisioning, deployment and incident response work, now in prev…
A developer released rag-debugger, an open-source MIT-licensed Python SDK that intercepts and inspects retrieval-augmented generation (RAG) pipelines to surface missing knowledge-base coverage. The tool, installable via …
A developer argues that multi-agent systems fail in production not because of model quality but because their orchestration layers lack the distributed-systems discipline needed for partial tool failures, context drift, …
A developer published the specification for a validation harness that checks agent configuration kits before release, verifying required files, a sub-300-line CLAUDE.md budget, parseable Cursor frontmatter, baseline sect…
Only 7% of organizations running AI workloads on Kubernetes deploy them daily, while 66% already deploy AI on Kubernetes and 47% do so occasionally, according to figures Apple Principal Engineer and CNCF Technical Oversi…