Microsoft Agent Engineering Certificate: Foundry, Copilot Studio and Agent 365

Dr. Aki Wijesundara

PhD in ML | Google AI Accelerator Alum

Manu Jayawardana

Exited AI Founder | Founder, TAI Labs

Build one enterprise agent on Microsoft, all the way to production.

Your company runs on Microsoft 365, and someone has asked you to ship an agent. You open Foundry, Copilot Studio, Agent Framework and Agent 365, and none of the documentation tells you which one you’re supposed to be using.

Here is the answer. Foundry builds the agent. Foundry IQ grounds it. Copilot Studio fronts it so business users can maintain it. Teams and M365 Copilot deliver it. Entra Agent ID and Agent 365 govern it. Most enterprise agent projects fail because they pick one surface and try to make it do all five jobs.

In 4 weeks you’ll take one agent through the whole pipeline:

Week 1: build a Foundry agent with the Agent Framework and deploy it from source

Week 2: ground it on a Foundry IQ knowledge base and publish it into Teams

Week 3: evaluate it properly, with routing accuracy, citation accuracy, cost and latency

Week 4: govern it under Entra Agent ID and Agent 365, and write the readiness brief

Four graded weekly assignments, each shipping a working piece plus the reasoning behind it.

What you’ll learn

Take one agent from Foundry to Teams and Microsoft 365 Copilot, evaluated and governed, in 4 weeks.

Agent Framework 1.0: agents, tools, structured outputs
Hosted Agents with per-agent Entra ID and built-in content safety
Deploy from source with azd, not from a portal

Foundry IQ knowledge bases over messy real document sets
Citations, and refusal when the answer isn’t there
Handle contradictory and outdated sources

Foundry for code, Copilot Studio for what business users must maintain
Front-door routing and agent-to-agent delegation
When computer use beats an API

One-click publishing to Teams and Microsoft 365 Copilot
How tenant policy flows through, and how to test it
Model choice, including Claude Sonnet 5 and GPT-5.5 Chat

Golden set of 50+ real cases with adversarial and unanswerable inputs
Routing accuracy and citation accuracy measured separately
Cost per request, p95 latency, and automated regression

Entra Agent ID, Agent 365 posture, the agent registry
Approval gates on write actions, data residency
A handover pack with named owners

Learn directly from Aki & Manu

Dr. Aki Wijesundara

AI Founder | Educator | Google AI Accelerator Alum

Manu Jayawardana

AI Advisor | Founder, TAI Labs

Previous Students from

See all products from TAI Labs

Who this course is for

The enterprise engineer Asked to ship an agent inside a Microsoft shop and unsure which surface owns which job.
The solutions architect Needs a defensible reference architecture and a governance story for an internal review board.
The technical lead evaluating the stack Deciding whether Microsoft’s agent platform can carry a production workload, and what it costs

Prerequisites

Working knowledge of Python or C#
You’ll write and deploy agent code. This is an engineering course, not a low-code overview.
Access to an Azure subscription and a Microsoft 365 tenant
You must be able to deploy to Azure and publish to Teams. Setup instructions come in pre-work.
Around 7 hours a week
Three hours live, four hours building. Every Friday has a shipping deadline.

What's included

Live sessions

Learn directly from Dr. Aki Wijesundara & Manu Jayawardana in a real-time, interactive format.

8 live sessions

Two 90-minute hands-on sessions a week for 4 weeks. Every session opens with a stack-agnostic engineering principle, then its Microsoft implementation.

4 graded weekly assignments

Each one ships a working piece of the system plus the written reasoning behind it. Graded on engineering judgment, not product familiarity

6 enterprise scenarios with document sets.

Realistic corpora with deliberate contradictions and gaps, hand-labelled golden sets, and target metrics.

Reference build

A complete worked example taken through every stage of the pipeline.

Starter repo

Three hands-on hours with Aki and Manu, who build and ship with these tools every week.

Governance pack templates

Entra Agent ID model, Agent 365 checklist, approval matrix, handover doc.

Tenant setup clinic

A live session before the cohort so nobody loses Week 1 to access problems.

Weekly office hours

Deployment and tenant problems get priority.

Lifetime access

All recordings, repos and templates, plus a future cohort.

Certificate

Issued on shipping the deployed agent, evaluation report and governance pack

Maven Guarantee

Your purchase is backed by the Maven Guarantee.

Course syllabus

7 live sessions • 4 lessons • 4 projects

Week 1

Sep 16—Sep 20

Build in Foundry

2 items

Sep

LIVE Session: Build in Foundry

Wed 9/168:00 PM—9:30 PM (UTC)

Week 2

Sep 21—Sep 27

Sep

Optional: Office Hours

Tue 9/229:00 PM—10:00 PM (UTC)

Optional

Ground and connect

2 items

Sep

LIVE SESSION: Ground and connect

Wed 9/238:00 PM—9:30 PM (UTC)

Schedule

Live sessions

3 hrs / week

Three hands-on hours. You'll get grounded in how the Microsoft agent pipeline fits together, then build one agent step by step: surface decision, Foundry build, Foundry IQ grounding, a Copilot Studio front door in Teams, and a first evaluation pass. Bring a real agent need and leave with it half-built, not just notes.

Wed, Sep 16
8:00 PM—9:30 PM (UTC)
Tue, Sep 22
9:00 PM—10:00 PM (UTC)
Wed, Sep 23
8:00 PM—9:30 PM (UTC)

Weekly Assignment

3 hrs / week

Every Friday you ship a working piece of the pipeline plus a one-page decision record: Week 1 a deployed Foundry agent and tool design, Week 2 a grounded agent in Teams with a refusal spec, Week 3 an evaluation report with golden set and changelog, Week 4 a governance pack and readiness brief. Graded on engineering judgment, not product recall.

Async content

1 hr / week

One hour a week of async lessons between sessions: the pipeline in one diagram, GA versus preview, deploying from source with azd, when computer use beats an API, handling contradictory and outdated sources, agent-to-agent delegation and model choice, Application Insights and cost, approval gates and data residency. Watch before each live session.

Testimonials

The AI training approach is outstanding. Our team learned to build practical AI solutions that we could implement immediately in our educational platform. The hands-on methodology made complex AI concepts accessible to our entire development team.
Kavi T.
CEO of Tilli Kids / Stanford PhD
Not only are the instructors experts in their field, they're incredibly skilled at breaking down complicated AI concepts so students can grasp them quickly. Anyone interested in building foundational AI knowledge should take this training - it's worth the investment.
Dr. Elizabeth Creighton
Founder & Principal at Brazen
The instructors help break down AI model development and clearly have plenty of experience to help others learn about complex concepts like infrastructure setup. The practical approach to NLP and LLM applications was exactly what our team needed.
Alissa Valentine
NLP & LLM Real World Data Scientist
I sent my team through this training for upskilling, and the results have been remarkable. Within weeks, they became much more efficient at building automations and deploying AI agents at work. This program bridges the gap between theory and practice and it’s had a real impact on our productivity.
Aamir Faaiz
CEO of Bayseian