MLOps news and analysis on Web Pulse: 3323 curated articles tracking the latest MLOps developments, tools, and research, updated continuously from vetted sources.
A prosumer running four AMD R9700 GPUs on an AM4 platform inside a Proxmox VM with a PLX switch got peer-to-peer inference working, reaching 82GB of host RAM usage, roughly 110GB of VRAM and a 500k-token GPU KV cache. Af…
A developer argues that every tool connected to an AI agent should be treated as both a token cost and a permission grant, and proposes starting each agent from a minimal tool inventory. Citing Anthropic's example of fiv…
A developer built a Kaggle benchmark of 336 synthetic Grafana-style monitoring panels to test whether vision models can read dashboards the way an on-call engineer does, grading each model on five fields: event type, sta…
Hopsworks is hosting two self-hosted decision models, qwen and kumo, that play the procedural game Hops Run live at game.hopsworks.ai, with each move sent to a Hopsworks model deployment that returns a probability per mo…
Databricks published a reference architecture for real-time e-commerce product recommendations built on its Data Intelligence Platform, based on an implementation for an Asian fashion retailer serving over 1 million mont…
Yann Gilliot, founder of GiLabs, described a three-layer production architecture in which AI systems only propose outputs — OCR, language models and extraction tools attach confidence scores to every proposal — while a d…
A developer published a walkthrough for building a production-grade customer support triage pipeline in n8n that normalizes ticket payloads from Zendesk, Intercom, and email, extracts urgency, intent, sentiment, language…
A developer behind the Ranex project argues that test results are only meaningful when bound to the exact commit they observed, introducing a concept called "subject-bound evidence." Ranex materialises the committed tree…
A developer built Chez Robot, a recipe assistant agent that uses MongoDB's native Vector Search to handle both semantic retrieval and persistent agent memory, avoiding a separate vector store. The tutorial wires TheMealD…
A developer built feedback-memory, a roughly 200-line MCP server backed by local PGlite Postgres with pgvector that stores past approvals, edits and rejections along with the reasons behind them, then returns the most si…
Roboflow published a walkthrough of turning vision-language model predictions into a production inspection system, using GPT-6 Astra's detection of hot spots in thermal images of solar panels as the example. The post det…
Permit.io outlined the minimum evidence package required for an AI agent audit trail, arguing that LLM observability logs capturing prompts, tool names, and latency are insufficient for auditors. The company contends tha…
A Korean middle-school digital tutor built a small web app called Clipboard Share using Claude Code, then handed its growth loop to an AI agent governed by an intent file, deterministic metric bands, and CI auto-merge ga…
A four-tier intent-routing cascade ordered by cost, with each tier allowed to abstain and every exit leaving a replayable log record, can replace a single frontier-model llm.classify(query) call for a large share of traf…
AOTInductor, PyTorch's ahead-of-time inference backend compiler, emits Triton kernels as cubin files stored in the pt2 zip archive, which developers can unzip and disassemble with nvdisasm or cuobjdump --dump-sass to ver…
Researchers introduced Incident-Arena, a human-built benchmark of 20 tasks that deploy real-world open source production applications to ephemeral Kubernetes clusters, inject faults from the config layer through underlyi…
A developer building a slopsquatting detector removed three false-positive exceptions from its package-risk classifier after finding they cleared packages without real evidence. The rules treated npm provenance and PyPI …
A developer built an open-source time-travel debugger for AI agents that records execution traces, lets users rewind to earlier states, modify decisions, and replay or branch from any step. The project pairs a Python Age…
A data engineering analysis argues that reliable enterprise text-to-SQL depends less on model choice than on the context built around the model, citing production write-ups from Uber and LinkedIn and a dbt Labs benchmark…
Stanford's Scaling Intelligence lab released KernelBench, a benchmark and toolkit for evaluating whether large language models can generate correct and efficient CUDA kernels from PyTorch programs, with the HuggingFace d…