The @DataRobot Community Slack is now open :) If you're building with AI and want direct access to us, this is the place. We have the best AI researchers, ML engineers and agent builders up in here. Come ask something hard. 👇 join.slack.com/t/datarobot-c…

AI can flag a supply chain risk. But what happens next? The real value comes when AI can help teams act on the problem. @ScottwLuton and Kim Reuter talk with Vika Smilansky and Marc Amarillas of @DataRobot about using AI agents to tackle supplier risk, tariffs and operational disruptions, including a $2.5B risk addressed in 72 hours. 🎥 Watch on demand: hubs.ly/Q04xP9vy0

Most agents get tested once, by whoever built them, on prompts that work. No signal on a malicious payload or six turns of pushback. Agent Assist now red-teams agents before deploy: multi-turn attacks, fixes you approve, a readiness verdict for the PR. datarobot.com/blog/adversari…

Most AI pilots never make it past the test phase. What happens when you actually put AI to work? Launched today: @ScottwLuton, Kim Reuter, Vika Smilansky and Marc Amarillas of @DataRobot break down what it takes to move agentic AI from pilot to production, including a real $2.5B supply chain risk resolved in 72 hours. Watch the new on demand webinar: hubs.ly/Q04w6B_W0

Elvin Aghammadzada, DataRobot, mapped where a context window stops working: up to 40% full it's the smart zone, past that it's the dumb zone here's how to shrink an agent's context, from scratch: step 1 → measure your own edge, 2026 measurements put the start of degradation at 25% of the window, not near the end - a window is not a tank, it's a runway step 2 → count what's burned before the user says a word, an agent wired to 15 MCP servers spends over 100,000 tokens on tool definitions alone step 3 → load on demand, a skill's front matter is under 100 tokens and the body only arrives when it's needed, roughly 10x cheaper in context than plain MCP step 4 → put teachability on the buying checklist next to security and compliance, skills already ship natively in 26+ platforms step 5 → prune the junk, model-written skills perform worse than human-written ones and there is still no verification layer for them on August 1-2 OpenAI showed Astra: ten problems open for decades, cracked by coordinating multiple agents over long runs, with about $2,000 of tokens for all ten most people bolt on one more MCP server - he switches off everything the current step doesn't need bookmark & watch - this 26-min talk, then read the full context rot breakdown below ↓

Our own Jyothi Nookula breaks down why an agent failing in production isn't a platform problem or a developer problem. It's a secret third thing😎

Ready to move #AI agents from concept to production? Join @DataRobot & @nvidia on August 26 for an executive briefing & hands-on lab covering secure deployment, governance & enterprise-scale AI. Save your spot: carah.io/DataRobotWebinar

We've been asking our devs: who's actually responsible when an agent fails in production? The developer, the platform, or the org? One dev's perspective:

Not to name names 🫣 but who's actually responsible when an agent fails in production? The developer, the platform, or the org? One dev's take:

Join @DataRobot and @nvidia on 8/26 for an interactive workshop focused on building enterprise-ready AI agents from the ground up... Register here: carah.io/DataRobotWebinar

Dev to dev: your agent isn't broken. Your testing process is. The teams making it to production are the ones systematically catching errors, turning them into tests, and iterating before they become production fires. One DataRobot dev built a framework for exactly that:

Trustworthy AI should give similar answers to similar questions. If your agent's responses vary wildly, you've got a stability issue. Hear about how we've been tackling the stochastic nature of language models head-on:

You made an agent in 20 minutes...but spent 3 days getting it to production. If that hit a nerve, give Workload API a chance piped.video/3iV3obcLc-U?si=XRJp…

The enterprise AI story is changing. The first wave was chatbots. The next wave is AI systems that can actually run parts of a business. In our conversation with @Saha_Deban, CEO of @DataRobot, he explains why this shift changes everything, and what companies are getting wrong about enterprise AI. Listen to the latest episode of Nebius for Startups podcast: nebius.com/podcast

Switching AI models mid-project used to mean switching your whole coding agent, and starting a new vendor review. DataRobot OpenCode makes it a slash command instead. One agent, many model options, one governed key. datarobot.com/blog/datarobot…

What are the four parts of a first-class agent identity? That's a hard math problem... let our latest @DataRobot blog break it down for you datarobot.com/blog/what-a-fi…

Your agent shipped with your API key. Now it has all your permissions, you can't tell the two apart in your logs, and the only off switch breaks your own access. @datarobot is launching a new series on agent identity, check out part 1! datarobot.com/blog/your-iden…

A decade of open source at DataRobot, from KDD Cup code drops in 2014 to shipping the actual infrastructure agents run on today. Dive in here: datarobot.com/blog/a-decade-…

Packed house last night at @SandboxVR for AI Engineer World's Fair week 🕶️ 200+ AI builders, VR, good bevs and even better chats. Thanks @nebiusai @tavilyai @CopilotKit for co-hosting!

In SF for AI Engineer World's Fair? Join us and @nebiusai, @tavilyai and @CopilotKit for a happy hour with an immersive twist 👀 Grab a spot 👇 luma.com/sandboxvr

Agents are only as useful as the capabilities they can reach. Wiring in every tool by hand doesn't scale. DataRobot now supports Agentic Resource Discovery (ARD), so our Skills and MCPs are discoverable through a standard AI catalog. datarobot.com/blog/datarobot…