I built an AI platform that resolves each IT ticket in under 30 seconds. Now I'm teaching you to build agents like it.
A hands-on program. Build a working AI agent and take it to production in 4 weeks, with direct access to me every step of the way.
Join the waitlistEnrollment for this cohort has closed.
What you leave with
This is not a course you watch. It's a build you finish.
- A working AI agent, deployed and running in production, not a demo
- The architecture behind multi-agent systems: routing, specialization, tool use
- The judgment to know what belongs in a prompt, what belongs in a knowledge base, and why that separation matters
- A repeatable method you can apply to the next agent, and the one after that
Who it's for
- Engineers who want to build agents, not just prompt chatbots
- Companies who would rather upskill one engineer than buy one more tool
- People who see where this is heading and want to be on the building side of it
Not required: prior AI or machine-learning experience. Required: you can write code and you want to ship something real.
How it runs
- October 4 Cohort kicks off. Self-paced core concepts on your schedule, framework choice, you pick your build target. Live kickoff call to meet the cohort and get aligned.
- Week of October 6 Self-paced + weekly call. Build your first working prototype, live group call to demo progress and troubleshoot together.
- Week of October 13 Self-paced + weekly call. Add production safeguards: memory, observability, graceful failure.
- October 18-19 The intensive weekend. Two full days, hands-on, direct guidance from me, building until your agent is genuinely production-ready.
Plus: an ongoing group Q&A channel for the full 4 weeks, so questions get answered daily, not just on call days.
Reserve your seat
Sponsoring a team member? Companies can pay by invoice. Email admin@aitechpal.com with your company name, billing contact, and the engineer's details, and we'll send a Stripe invoice with payment terms.
About Your Instructor
I'm Ola, founder and builder of AI Tech Pal, a production multi-agent AI platform that integrates with ServiceNow, Jira, Zendesk, and Freshservice to resolve real-world IT support tickets in under 30 seconds.
I didn't just design the concept. I built and shipped the entire platform myself, from the AI agents and orchestration pipeline to integrations and production deployment.
I've spent 14 years in Enterprise IT, with experience spanning Dell, GEICO, and consulting engagements through the University of Maryland Medical Systems, the last 10+ of them in Systems Engineering, working in complex environments where reliability, security, and solving real business problems are non-negotiable.
I also bring years of experience organizing training programs for startups and entrepreneurs, so teaching people how to turn ideas into something practical isn't new to me.
This isn't a theory-first AI course. I'm giving you the same practical playbook I used to go from an idea to a working, production AI system, and we'll build hands-on until your own AI agent is live and working.
You won't just learn about AI agents. You'll leave having built one.
Questions
Do I need AI experience?
No. You need to be comfortable writing code. We cover the agent-specific concepts together.
What will I actually build?
A working agent suited to your goal: a personal-workflow agent, a small-business operations agent, or an internal-task agent for a company. You pick the target in week one.
How much time should I budget?
The core content is self-paced, budget 2-3 hours a week to go through it. Plus one weekly live call and the full intensive weekend.
Is this still live and personal, or am I just watching videos?
Both. Core concepts are self-paced so you're not boxed into a fixed schedule, but you get a live group call every week plus a daily-monitored Q&A channel and the full hands-on weekend with me directly.
What tools and stack?
Python, plus Microsoft AutoGen for agent orchestration, the same stack behind AI Tech Pal in production. You bring an API key from any major LLM provider (OpenAI, Anthropic, Google, or an open-weight model you host yourself); the architecture works the same either way. The principles you learn transfer to other frameworks like LangGraph or CrewAI, and we'll cover when each one fits.
Is the mentorship really until it's live?
Yes, bounded to the 4-week window, which is designed to get a committed builder to production.
Refund policy?
Full refund up to and including October 7. After that, seats are capped at 12 and can't be refilled, so no refunds. But your seat isn't lost: you can transfer it to the next cohort or to a colleague at your company.
Can my company pay?
Yes, by invoice. See the pricing section above.