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😎
1
2
620
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:
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:
490
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:
589
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
youtu.be/3iV3obcLc-U?si=XRJp…
2
908
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!
Skills as SDKs. 🗣️
DataRobot agent skills install into Claude Code in one command and teach it your platform's real conventions, so it stops guessing and starts shipping.
Live churn-scoring demo in the video 👇
youtube.com/watch?v=v-cR1vfP…
What if your agent didn’t need a prebuilt tool registry?
In our latest @DataRobot Build Club, @KindofLuke builds an agent that reads an OpenAPI spec, writes its own tools, tests them in a secure sandbox, and turns the working code into a deployed agent.
youtube.com/watch?v=w9jocCUI…
3
331
A digital twin that triages your Slack, drafts replies in your voice, and only escalates what needs you. Built in an hour on @DataRobot!
The demo isn't the point: moderation, PII guardrails, and observability are what make it defensible:
youtube.com/watch?v=JVxVrv8k…
1
3
334
DataRobot skills are now in the Cursor marketplace.
Build, test, and ship an AI agent without leaving your IDE: setup DataRobot with our skill, scaffold an agent, spin up an MCP and deploy to prod with secure infra handled for you. Check it out 👇
youtu.be/opAqpoOKlT8?si=TToz…
1
4
591
One dev's take: the root problem is almost never the agent. It's the lack of guardrails designed to catch errors before they escalate:
3
271
What pain points are developers feeling that nobody's talking about?
Dev to dev: the honest answer is disorientation. 🧭
1
301
Dev to dev: the developers who pull ahead won't be the best prompters. They'll be the ones who can think in terms of business value, communicate clearly, and use AI to interrogate their own ideas — not just generate them.
That used to be senior-level work. Now it's the job description. One dev's take:
1
3
240
Stuck on an eval framework? Wrestling with agent orchestration? Debugging a prod model at 2am? Our Slack has been there. Join us. → join.slack.com/t/datarobot-c…
2
160




