Automate evaluations | Microsoft Foundry

Build AI agents that meet your standards for quality, safety, and performance using Microsoft Foundry. Trace every run end-to-end, generate synthetic datasets to stress-test on demand, fire automated Red Team attacks at your own agents, and pin down why evaluations fail — all from the Microsoft Foundry control plane. Lock in guardrails that inspect every tool call at runtime, define the risks once, and enforce them across every agent run.

Mohammad Abuomar, Responsible AI Principal Architect, shares how to turn a coding agent into production-ready software inside Foundry.

► QUICK LINKS:

00:00 - Microsoft Foundry control plane

00:33 - See a finished agent

02:30 - See where the agent started

03:19 - Traces

04:04 - Built-in monitoring

04:34 - Evaluation types

05:51 - Red team evaluations

07:08 - Evaluation results

08:14 - Built-in Guardrails

08:14 - Wrap up

► Link References

Get everything you need in Microsoft Foundry at https://ai.azure.com

► Unfamiliar with Microsoft Mechanics?

As Microsoft's official video series for IT, you can watch and share valuable content and demos of current and upcoming tech from the people who build it at Microsoft.

• Subscribe to our YouTube: https://www.youtube.com/c/MicrosoftMechanicsSeries

• Talk with other IT Pros, join us on the Microsoft Tech Community: https://techcommunity.microsoft.com/t5/microsoft-mechanics-blog/bg-p/MicrosoftMechanicsBlog

• Watch or listen from anywhere, subscribe to our podcast: https://microsoftmechanics.libsyn.com/podcast

► Keep getting this insider knowledge, join us on social:

• Follow us on Twitter: https://twitter.com/MSFTMechanics

• Share knowledge on LinkedIn: https://www.linkedin.com/company/microsoft-mechanics/

• Enjoy us on Instagram: https://www.instagram.com/msftmechanics/

• Loosen up with us on TikTok: https://www.tiktok.com/@msftmechanics

Tämä jakso on lisätty Podme-palveluun avoimen RSS-syötteen kautta eikä se ole Podmen omaa tuotantoa. Siksi jakso saattaa sisältää mainontaa.

Jaksot(100)

Agent 365 | Security Operations in Defender

Surface every AI agent in your tenant and expose the ones throwing security signals — across both the IT and SOC view. Triage high-severity alerts as IT in the Microsoft 365 admin center, then pivot i...

29 Touko 7min

Microsoft Entra Tenant Governance | Find Configuration Drift

Ensure your tenant configuration doesn't drift from defined security and compliance requirements with Microsoft Entra Tenant Governance. Capture configuration as code across 200+ resource types in Ent...

27 Touko 8min

Microsoft Excel Beginners Tutorial (2026)

This is the Microsoft Excel guide and tutorial for beginners. If you're new to and getting started with Excel or coming from another app, in this video we teach the basics of Excel, the user interface...

18 Touko 12min

Work IQ | Data, Context, Skills & Tools for Copilot and Your Agents

Ground every Microsoft 365 Copilot response in your real work data. Pull context from SharePoint, OneDrive, Teams, email, and meetings — all through Work IQ. Draft Word documents that carry your exist...

13 Touko 9min

Azure Arc | On-prem + Multi-cloud Management

Managing Servers, and Kubernetes across on-prem, and multiple clouds, can quickly become complex, especially when you're juggling multiple tools. In this video, we explore how Azure Arc simplifies hyb...

8 Touko 14min

Agent 365 | Your Security & Compliance Controls

Keep AI agents operating under control and within your data security and compliance policies using Agent 365. Block agent access to labeled files at runtime, stop sensitive data from leaving in agent-...

7 Touko 8min

Operations Context for AI | Ontology in Fabric IQ

Build a shared semantic foundation for your enterprise data in Microsoft Fabric IQ. Generate a full business ontology from an existing Power BI semantic model, map entities and relationships, and embe...

28 Huhti 9min