MLOps news and analysis on Web Pulse: 1312 curated articles tracking the latest MLOps developments, tools, and research, updated continuously from vetted sources.
Prefill/decode disaggregation, which runs LLM prefill and decode phases on separate hardware pools, cut p99 inter-token latency by 66% (from 88ms to 30ms) in a Go simulation, at the cost of higher time-to-first-token (42…
A developer detailed the deployment of Google's Gemma 4 2B model on a single Cloud TPU v5e chip using MCP and Antigravity CLI, achieving measured throughput and latency. The v5e chip, at roughly half the cost of v6e, pro…
A developer outlines a cost-reduction strategy for LLM API bills in SaaS applications, advocating for prompt routing to smaller models with fallback to larger ones, and batch processing for delay-tolerant tasks. The appr…
A developer reviewing an AI agent platform for automated video generation found that the architecture's practice of giving every agent nearly the entire execution context creates significant production liabilities. The e…
Karpenter best practices from CAST AI recommend designing multiple focused NodePools with taints for workload isolation, enabling the SQS interruption queue for Spot, and setting CPU and memory limits on every NodePool t…
According to the Cast AI 2026 State of Kubernetes Optimization report, which analyzed 23,000+ clusters, the median CPU utilization across autoscaled EKS clusters is 8%, with 69% of Kubernetes clusters over-provisioning C…
A developer argues that the real challenge in AI is not building the agent but the infrastructure around it, including tool integration, memory, hybrid search, and observability. The post highlights that production AI sy…
A developer at Programming Tech Lab explains linear regression using a car valuation example, demonstrating how the algorithm fits a line to data and predicts prices based on mileage. The post includes a Python code samp…
A developer outlines an LLM latency budget framework for AI SaaS builders, emphasizing that slow AI features feel broken and that the fix requires product-level constraints rather than simply using faster models. The app…
AWS launched Kiro Crew, an open-source orchestration platform designed to enable autonomous engineering workflows by coordinating multiple AI agents across repositories and developer tools. The system, previously used in…
An engineer's benchmarking guide introduces a cache hit rate metric for measuring prefix caching effectiveness in LLM serving, implemented in the open-source tool llmperf-rs. The metric calculates the ratio of cached tok…
A developer audited their coding agent's to-do list and found that 27% of open tasks were already completed, revealing a systemic flaw in how AI agents report task completion. The developer's setup uses GitHub issues as …
TIER IV, the Tokyo-based developer of the open-source autonomous driving software Autoware, and automotive supplier Astemo signed a memorandum of understanding on August 5, 2026, to jointly build a next-generation develo…
Perplexity is hiring a Member of Technical Staff (Software Engineer, GPU Cluster Infrastructure) in San Francisco with a salary range of $250k–485k/yr, to build a self-serve compute platform for its GPU fleet across mult…
Mastercard is hiring an AI Engineer for its Global Business Services Center in Pune, India, to design and implement AI and Generative AI solutions using LLMs, Agentic AI frameworks, and RAG architectures. The role involv…
A developer has released Wide Events, a Rails gem that attaches production context to a single OpenTelemetry root span per request or job, enabling coding agents to quickly diagnose issues. In one example, the root event…
Mistral AI is hiring a Research Engineer for Code Agents Infra in Palo Alto, California, to build and operate the infrastructure powering its agentic models and coding assistants, including sandboxing for over 1 million …
Hyperspell, the memory and context layer for AI agents, uses Firetiger's agents to automate production monitoring and fixes, catching Kubernetes crash loops and cleaning up Temporal workflows. Platform engineer Charlie S…
Cloudflare has introduced the Agent Development Lifecycle (ADLC), a new framework and set of tools designed to let AI agents manage the entire software development lifecycle, not just code generation. The company announc…
Perforce Software's 2026 Platform Engineering Report, based on a survey of 820 technology professionals, found that 73% of organizations with mature platform engineering practices said platform maturity was critical or s…