15 stories
Learn how agent traces, human oversight, production evaluations, and log retention support EU AI Act compliance with a practical credit-assistant example. The post EU AI Act compliance for AI agents: a practical engineering guide appeared first on Arize AI.
Evaluating OpenRouter? It routes model calls across providers. Vivgrid runs enterprise agents — every model on one key, cloud-hosted tools, and evals.
Automated evals can score every agent session, but they do not automatically reflect what your team considers a good outcome. That takes human judgment. The post Alignment is a human-AI workflow: how Custom Views make agent review possible appeared first on Arize AI.
AI costs don’t always scale predictably with traffic. A retry loop, oversized context window, or unnecessarily expensive model can make a small number of traces disproportionately expensive. The post Cost Agent: find where your LLM and agent spend goes with your traces appeared first on Arize AI.
This year agents at OpenAI, Anthropic, and Google escaped environments meant to contain them. What decided severity was time-to-detect and time-to-kill, from 12 minutes to seven months. NVIDIA's Open Agent Safety Platform puts a kill switch in hardware the agent can't reach.
Agents built and deployed on MongoDB’s Atlas Agent Engine can export OpenTelemetry traces to Arize AX, where teams can inspect each run, evaluate both the outcome and execution path, and test fixes before redeploying.
Sessions are now a first-class unit of work in Arize AX. Annotate a whole conversation, queue it for review, and ask Alyx to filter for it — plus Agent-as-a-Judge on every plan, vision judges, faster datasets, and everything else that shipped between August 6 and September 18, 2026.
Evaluating Baseten? Here's an honest comparison: Baseten is an inference platform for deploying and scaling your own models. Vivgrid is where enterprise agents run — frontier and open models on one key, cloud-hosted tools, evaluation before release, and BYOC that routes sensitive requests to your…
Evaluating Fireworks AI? Here's an honest comparison: Fireworks is a training and inference cloud for open-weight models. Vivgrid is where enterprise agents run — frontier and open models on one key, cloud-hosted tools, evaluation before release, and BYOC that routes sensitive requests to your own…
Evaluating Together AI? Here's an honest comparison: Together is an AI-native cloud for running and fine-tuning open-source models. Vivgrid is where enterprise agents run — frontier and open models on one key, cloud-hosted tools, evaluation before release, and insight into what customers actually…
We finished 11th on Product Hunt with organic votes — and the thoughtful questions about observability, cost, loops, and human control confirmed Mozaik's roadmap toward controlled autonomy.
A best-practices guide for enterprise AI agents: treat system prompts as runtime config, run tools as Managed Skills, and make models a platform decision — so you can debug faster, ship AI improvements without app releases, and fix production behavior from the console in minutes.
Personal agents run tools locally. Enterprise agents can't — credentials, auditing, consistency, and iteration speed all break at company scale. The answer is Managed Skills: agent tools that run in the cloud.
Production logs aren't observability for multi-agent AI. kaleidoskop replays every Mozaik orchestration as a scrubbable hex grid — bus events ripple, agents light up, you find the moment that broke the run in seconds.
Automate customer emails from your AI agents using Autosend MCP on Vivgrid. Generate and send transactional emails, notifications, and more — no code needed.
Showing 15 of 15 stories
You're all caught up.
Get the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts
