agentic ai · course & live cohorts
Put AI to work in your real workflows.
A hands-on course for professionals who want AI doing useful work, not just answering questions. Unlike a chatbot that returns one answer, an agentic workflow moves through a process, uses approved tools, and pauses for human judgment when it matters. You'll map a real process, configure it in a guided learning platform, run it against realistic sample data, and evaluate the results. No prior coding experience required.
workflowsagents-and-toolsevaluationgovernance
1. read new invoices from inbox ✓2. extract vendor, amount, due date ✓3. flag anything unusual → 2 need review4. draft approval emails ✓5. wait for your sign-off ● waiting
paused · nothing sends without a human
what you'll learn
From "AI is interesting" to AI doing real work
This isn't a prompt-engineering course. You'll design AI into real work, with the oversight serious organizations actually require. Every workflow starts by hand, then with AI, so you see exactly where it helps and where your judgment still rules. You practice on realistic sample data, so nothing sensitive from work is ever required. Enrollment is planned to open in fall 2026. Exact cohort dates and the final lesson sequence are still being finalized. The waitlist hears first.
You'll start with working templates and adapt them step by step using readable, structured configuration rather than application code. The platform runs in a browser or locally, so you can see each workflow execute and inspect what happened.
- Designing AI-enabled workflows: break a real business process into steps AI can genuinely help with.
- Agents, tools, and orchestration: what these words actually mean, and when each is the right choice.
- Instructions that hold up: write prompts and process definitions that produce more repeatable, testable results instead of one-off demos.
- Evaluating quality: judge AI output, catch failures, and tell whether a workflow is getting better or worse.
- Guardrails, governance, and oversight: set boundaries on what AI may do, keep a human in the loop, and leave an audit trail that proves what happened.
- Data you can trust it with: decide what company information is safe to give an AI tool, what to keep private, and how to handle sensitive data responsibly.
- Security and safety: recognize confidently-wrong answers, spot manipulated inputs that try to hijack what the AI does, and set limits on what it may act on by itself.
- Working alongside AI: practical habits for delegating to AI without losing sight of accountability.
You'll leave with a portfolio of working workflow prototypes, plus the judgment to decide where AI belongs, where human review is required, and when AI is not the right tool.
These are functioning prototypes built with sample data, not production systems connected to your employer's tools or information.
who it's for
Built for the people who own the process
A good fit if you…
- Run or manage business processes AI could help with
- Use ChatGPT-style tools and suspect there's much more
- Are accountable for the quality and risk of AI use at work
Probably not for you if you…
- Want to build production AI software from scratch: that's an engineering course
- Are looking for machine learning theory or the math behind models
What you need: A computer, an internet connection, and a modern web browser. Access to the guided course platform is included. You can complete the exercises in your browser, with an optional local setup if you want it. Final third-party AI-service account requirements and any usage costs will be shared before enrollment.
live cohorts
Self-paced when you want it. Live when it helps.
Cohorts are live, remote, and kept small: around 25 learners. You get structure, momentum, and a real person answering your real questions: the things that are hardest to get from a video library.
- Instructor-led sessions: live and remote, not pre-recorded.
- Hands-on exercises: you build the workflows during class, not after.
- Structured progression: each week builds on the last.
- Live questions: get unstuck in minutes, not days.
- Practical projects: finish with a working workflow you can point to.
- Feedback and peers: direct guidance, and classmates solving the same problems.
teaching approach
How the teaching works
Practical over theoretical
Concepts are introduced when you need them to build something, not as a syllabus to memorize.
Learn by building
Every exercise ends with a workflow that runs. Small wins compound into real capability.
Plain language
Agents, orchestration, evaluation: every term gets a working definition you can repeat to your team.
Skills that transfer
Exercises mirror actual work: messy inputs, changing requirements, and results someone has to stand behind.
instructor
About David Stanek
David Stanek is a software engineer and educator with nearly three decades of professional experience. As a Principal Cloud Software Engineer at HPE, he works with the technologies and practices that power modern organizations.
David uses agentic systems in real operations, not just demonstrations. He has automated multiple business processes across two businesses he owns, built custom AI-agent harnesses for other companies, and uses bounded agents to support personal workflows involving calendars, email, and commitments.
Since 2019, David has taught Python at Case Western Reserve University, including students without technical backgrounds. That combination of applied AI work and experience making technical subjects approachable shapes this course. He is also building the guided platform students will use to run and evaluate their workflow prototypes.
expertise: software engineering · cloud · automation · ai · teaching
pricing
Two ways to take it
Put AI to work your way. Work through the material on your own, or join a live cohort of around 25 learners for instruction and feedback on your actual workflows. The plain-language, no-code approach is the same either way.
self-paced
$199, regularly $299
launch price
- The complete course, yours to work through on your own schedule.
- Plan for about 25 hours of lessons, guided exercises, and projects. Take more time whenever you need it.
- Access to every lesson and example for as long as Learn, Fast continues to offer the course.
- Once enrolled, start whenever you like and learn at your own pace.
- A private Discord community where David generally responds to questions within one business day.
refund Full refund before you reach the 4th module.
live cohort
$1,899, regularly $2,499
launch price
Everything in self-paced, plus:
- 35 hours of live instruction in a small cohort of around 25 learners.
- Pick your schedule: once a week for 14 weeks, twice a week for 7 weeks, or four full Saturdays.
- Live Q&A and feedback on your own workflows as you go.
- Learn alongside your cohort in the private Discord community.
- Miss a session? Every one is recorded and available to your cohort for 30 days afterward.
- Keep access to the self-paced materials for as long as Learn, Fast continues to offer the course.
refund Full refund if it's not for you: before week 4 of the 14-week schedule, before week 3 of the 7-week, or before the 2nd Saturday of the four-Saturday intensive.
Launch pricing applies to the first enrollment window; the regular price takes effect after that. Cohort dates are announced to the waitlist first. Join the waitlist to hear when enrollment opens. Taking it with your team? Groups of five or more get 20% off every seat. Get in touch for a quote. Many employers cover professional development: you will get an itemized receipt and a course outline to share with your manager.
waitlist
Be first to hear when Agentic AI enrollment opens
Join the waitlist and you'll get course announcements, cohort dates, and early enrollment access for the Agentic AI course. Nothing else, and you can unsubscribe any time. Joining costs nothing and commits you to nothing.
Not ready to sign up? Ask a question first.
Pairs well with the Python course: same plain-language approach. Python is not required for this course, but it can expand what your workflows can do.