Agentic AI PM Sprint
Aki Wijesundara, PhD
AI Founder | Google AI Accelerator Alum
Manu Jayawardana
AI Advisor | Co-Founder & CEO at Krybe
Stop Building Agent Demos — Ship Agents You Can Trust in Production
Turn agentic AI from experimental prototypes into reliable, production-ready systems. Most agents fail silently in production—broken tool calls, infinite loops, partial task completion, and unpredictable results. The problem isn’t just prompts; it’s the lack of a repeatable system PMs can own.
Agentic AI PM Sprint teaches PMs to design, control, and measure agents that actually complete tasks. You’ll:
✅ Decide when an agent is the right solution versus workflows or simple automation
✅ Map the full agent loop: triggers, planning, tool calls, memory, stopping rules
✅ Define tool contracts (inputs, outputs, retries, permissions) to prevent failures
✅ Build fallbacks, safe mode, and human-in-the-loop escalation
✅ Set metrics that predict success and detect drift early
✅ Ship with staged rollout and monitoring plans PMs can defend
Week by week, move from uncertain prototypes to structured agent specs, measurable KPIs, and robust rollout plans. Teams using this approach cut failed launches 40–60%, reduce post-launch firefighting, and scale faster—replacing guesswork with control, reliability, and stakeholder trust.
What you’ll learn
You’ll create an Agent Spec Pack for a real agent you’re building.
Map the agent loop: triggers, planning, tool calls, memory, stopping rules
Decide when to use an agent, a workflow, or simpler automation
Translate desired outcomes into measurable success criteria
Identify why agents break in production and turn vague concerns into actionable fixes.
Create a failure taxonomy for agentic tasks and tool interactions
Separate leading indicators (tool success, retries, time-to-complete) from lagging indicators (task success, escalation rate)
Define inputs, outputs, error states, retries, and permissions for all tools
Build fallbacks, safe mode, and human-in-the-loop escalation paths
Run lightweight monitoring and weekly agent ops cadences
Who this course is for
Product managers building agentic AI who want reliable, production-ready agents, not experimental prototypes.
PMs with LLM and automation basics who want a practical, data-driven way to define behavior and measure success.
PMs shipping AI in regulated industries who want a repeatable, approval-ready process instead of slow cycles.
What's included
Live sessions
Learn directly from Aki Wijesundara, PhD & Manu Jayawardana in a real-time, interactive format.
Hands On Customized Resources
Get access to a customized set of resources
Lifetime Discord Community
Private Discord for peer reviews, job leads, and ongoing support forever.
Guest Sessions
Webinar sessions hosted with industry network
Certificate of completion
Showcase your skills to clients, employers, and your LinkedIn network.
Maven Guarantee
This course is backed by the Maven Guarantee. Students are eligible for a full refund up until the halfway point of the course.
Course syllabus
Week 1
Lecture 1: Foundations of Agentic AI
Week 2
Lecture 2: Agent Loops & Tool Contracts
Schedule
Live sessions
1 hr / week
Live sessions 6 hrs
6 Prerecorded Lectures 6 hrs
2 hrs / week
Short, focused videos that break down the complete AI evaluation framework, designed for quick learning and easy rewatching as you apply it in production.
6+ Office Hour Q&As 6 hrs
2 hrs / week
Open office hours for deep dives, debugging help, and personalized feedback.