Stay-at-home mom → Test Automation Engineer
Tifa
“My daughter started asking where the money comes from. Now she sees her mom working in tech.”
Read her story
Agentic QA for Professionals · 4-week live cohort
AI didn't take your QA job. It's turning into Agentic QA.
Now enrolling
Starts August 27, 2026Thursdays & Fridays11:00 AM – 3:00 PM ET+ Wednesday prep sessions
Where our graduates work
14,000+ engineers trainedAcross 36 countriesTaught live by the founderCohort 1 complete · 26.9 hours
The app you test
You practice on a real bank, not slides.
ZincBank runs in production. You point the agents at it, and you own the call on what they got wrong.
Proud member of the
Claude Partner Network
Claude Code is one of the agents you operate in the cohort. We build and teach with it as an official partner.
The honest part
AI writes the tests now. Someone still has to run the agents, and decide what “correct” means.
That someone points the agents at the app, owns the definition of “working,” and catches the money bugs before customers do. It's the one part of QA that grows because of AI. It's also the part you don't have yet.
What AI already does
80%
The typing. Agents draft the tests, wire the selectors, and file the first pass.
Agentic QA, today
20%
The judgement. Someone still points the agents, and decides what correct means.
Why this, why now
Three things change the moment AI enters your test suite.
Run the full agentic testing lifecycle, on a real banking project
Against Zinc Bank, a real (fully simulated) banking app, you drive the whole loop: read the requirement, author the test, execute it, heal what broke for the wrong reason, and grade the result. Not slides. The actual loop, on the actual app, with you deciding what counts as correct.
analyze→author→evaluate→execute→heal
the loop
Operate a real framework, and build the layer under it
A production-grade Playwright + BDD framework driven by planner, generator and healer agents over MCP — the same architecture Playwright now ships as first-party Test Agents. Then you go one level down and build your own MCP server, which is the part almost nobody else has done.
the framework
Win the interview every QA now sits
Where AI helps and where it silently fails, why a human still decides what 'correct' means, and how to talk about token, context and data governance without hand-waving. That 20% is now the whole interview.
the edge
The architectureSeven agents.
Seven agents.
One human who decides.
Most agentic-QA training stops at three agents, because that is what Playwright ships for free. You build the whole chain, including the two stages nobody else talks about.
Requirement Analyzer
Requirements
Reads the requirement as a QA, not a developer. Flags what is untestable before anyone writes code.
Test Case Writer
Authoring
Extracts the acceptance criteria and drafts the cases, citing the requirement ID for every one.
Test Case Evaluator
The difference
A second, fresh agent grades the first one's work before a line of automation gets written.
Test Script Writer
Automation
Turns approved cases into Playwright and BDD, against your page objects and fixtures.
Script Evaluator
The difference
Proves every acceptance criterion has a matching assertion. A passing test that asserts nothing is the bug you never find.
Executor
Execution
Runs the suite and heals what broke for the wrong reason — a moved element, not a real defect.
Investigator
Triage
Analyses the failures that are real, so you read findings instead of logs.
Traceability check
Across all seven
Every requirement mapped to its case and its script — proposed by an agent, verified by deterministic code.
An agent that writes tests is a demo. An agent that grades another agent's tests is a system you can put in front of a release.
The curriculum
10 live sessions. One real framework.
Four weeks: eight core sessions, plus the Wednesday interview-prep sessions. You leave with a framework you can defend line by line.
01
How the models actually fail
Why a model can be fluent and wrong at once, and what that changes about testing. Wire a CLI coding agent and MCP to Zinc Bank, then run a real requirements doc through the analyze · author · execute loop end to end.
02
MCP, and the framework under it
Stand up the Playwright + BDD framework — page objects, fixtures, data, the parts you'll be asked to whiteboard. Then build your own MCP server, so you understand the layer everyone else only consumes.
03
The evaluator layer
Add the two agents nobody else teaches: one that grades the test cases, one that proves every acceptance criterion has a matching assertion. Forced citation so nothing gets invented, and a traceability matrix checked by deterministic code.
04
Orchestration, cost, and the interview
Put the whole chain under one orchestrator. Build the activity inventory that maps each task to the cheapest model that can do it. Then rehearse the answers hiring managers now expect, out loud.
The interview
These are the questions. Can you answer them yet?
Eight of the 386 in the cohort's interview bank. Every answer traces back to a moment in a real class.
- An agent reports that all tests pass. How do you verify that claim?
- How would you review a thousand generated test cases when you cannot read them all?
- How would you design a definition of done that an agent cannot fake?
- Your company will not send data to a third-party API. What are your options?
- Why are some models dramatically cheaper than others when quality is close?
- The agent can write the framework. Who decides the architecture, and why does it matter?
- How do you verify a UI action really worked, rather than just looked like it worked?
- How would you detect a hallucination automatically?
There are 378 more. You leave with all of them.
Who's teaching this
Two teachers. Two layers.
Adam Bary builds the framework in front of you, out of live production work. In the last cohort that meant debugging a real migration of 6,000 test cases on screen, unscripted, because the number in the dashboard was wrong.
I'm Kuzzat Altay. I founded CYDEO, trained more than 14,000 engineers across 36 countries, and I teach the layer that decides whether any of it can be trusted: how the models actually fail, what an agent must cite before you believe it, and what each call costs. I also read every résumé myself.
What we teach here, we run in production.
Where I've built and studiedAccentureCGIInovalonHarvard Business SchoolPlanckVPN
Kuzzat Altay · founder
What we ship
We don't just teach this. We build it.
ZincBank, the app you test in the cohort, is one of several products my team designed, shipped, and runs in production.
Fintech · QA training
ZincBank
The production-grade banking app you point the agents at in class.
● Live · zincbank.cydeo.io
Privacy · Security
PlanckVPN
An independent, zero-log VPN built on WireGuard.
● Live · planckvpn.com
CRM · SaaS
Nuebase
The CRM that runs our whole cohort business.
● Live · nuebase.com
AI · Developer tools
CYDEO AI Lab
An AI coding assistant built for QA engineers, not just prompts.
● Live · ai.cydeo.com
Family safety
NURPN
A family-safe VPN and content blocker.
● Live · nurpn.com
Cybersecurity · SOC
Security Operations Center
A real SOC where you watch analysts work, taught hands-on.
● Live · cydeo.com/soc
Be honest with yourself
This isn't for everyone.
Come if
- You're a manual or automation tester ready to add the AI layer.
- You keep hearing “agentic,” “MCP,” “self-healing” in interviews and want to actually operate them.
- You're comfortable with Java/Selenium or Playwright, you already have the 80%.
- You learn by building on a real app, not by watching slides.
Skip it if
- ✕You want a six-figure job handed to you for little effort.
- ✕You expect the full six-month career path compressed into four weeks.
- ✕You won't put in reps between the live sessions.
What you keep
Four weeks of class. A permanent arsenal.
The sessions end. Everything below does not.
Interview prep
386 interview questions
Every answer traced to a moment in class, in plain language you can say out loud.
Reference
The study guide and glossary
74 topics ordered basic to advanced, plus the agentic SDET glossary and field manual.
Practice
A mock-interview platform
Rehearse the answers out loud, on your own time, as many times as you need.
Tooling
ZincTM and its MCP server
A real test-management system with your own account, so you own the whole workflow.
Code
The framework repo
The agentic framework itself, plus prompt templates for code review, DevOps, database and QA work.
Access
Every session recorded
Yours permanently, at no extra cost. Rewatch any session whenever you need it.
After the cohort
It does not stop at week four.
- Weekly Wednesday 7:00 PM ET sessions with the founder, running on after the cohort ends.
- A two-day bootcamp on building the framework from scratch.
- Ad-hoc sessions whenever the tooling shifts, announced in Discord.
- Your résumé read by the founder personally.
- A check-in at three months to hear where you landed.
What it costs
One flat rate.
Everything included: the live cohort, the résumé-ready framework, and $50 in real AI credits (Claude Code, Qwen, and DeepSeek) to run it.
Working QA professional
$1,299
- $1,249 tuition + $50 AI credit
- 10 live sessions over four weeks
- The framework repo, 386 interview questions and every recording, permanently
- The full agentic upskill for QA engineers, managers, and tech leads.
30-day money-back guarantee
Cohort 1
The first cohort finished on 14 August 2026.
10
live sessions
26.9h
of class, recorded
386
interview questions written from it
Who we train
We have done this 14,000 times.
Most of them started with no tech background at all. If we can take someone there to a QA job, we can take you from the 80% to the 20%.
Military veteran → Software Quality Assurance
Bradly
“The discipline from the military helped, but CYDEO gave me the skills. Completely changed my life.”
Read his story
Babysitter & odd jobs → QA Engineer
Husna
“When I saw my first paycheck, I couldn’t believe it. Nobody in my family thought seven months could change everything. It did.”
Read her story
Personal trainer → Automation Engineer
Tariq
“As a trainer, I could only count reps. Now I build test frameworks. It showed me how much potential I really have within myself.”
Read his story
Delivery driver → QA Engineer
Lilia
“From delivering packages to writing test automation. I didn’t think this was possible for someone like me.”
Read her story
2 AM shifts, no work-life balance → Cybersecurity Analyst
Yvette
“Everyone had their doubts. A friend called it a scam. I proved them wrong through hard work. Now I protect data from home with a cup of coffee.”
Read her story
Restaurant manager → SDET
Omar
“I was tired of standing. Eight months of study taught me what four years of college couldn’t. Three raises in my first year without even asking.”
Read his story
Nail tech & cab driver → Automation Engineer
Miriam
“I used to drive passengers to the buildings where I work now. I wasn’t smart, but I had desire. Now I’m making six figures and never missing my daughter’s life.”
Read her story
Questions
Before you join.
When does it run?
10 live sessions over four weeks, starting Thursday, August 27, 2026. The eight core sessions run Thursdays and Fridays, 11:00 AM to 3:00 PM ET: Aug 27-28, Sep 3-4, Sep 10-11, and Sep 17-18. On top of those, Wednesday 7:00 PM ET interview-prep sessions run with founder Kuzzat Altay, and they keep going after the cohort ends. Every session is recorded and stays yours.
Is enrollment open right now?
Yes. You can enroll and pay today, and your seat is confirmed straight away. This is the second cohort, starting Thursday, August 27, 2026, and seats are capped at 60. If you would rather not commit yet, the form lower down puts you on the list for the cohort after this one.
What do I need to know already?
It's an upskill for working SDETs. If you're comfortable with Java/Selenium or Playwright and basic test automation, you have the 80%. We build the agentic 20% on top.
What's the $50 token credit?
Every seat includes $50 in AI token credits for Claude Code, Qwen, and DeepSeek, the models you run in class. You practice on those credits instead of paying out of pocket.
Why isn't the upskill just free?
Because paid students finish. In our experience, people who pay anything complete at far higher rates than a free cohort, roughly 80% higher for us. With the $50 token credit built in, your real tuition is almost nominal. The small commitment is what carries you to the end.
Which AI models do we use?
Claude Code is the agent you operate most, and your credit also covers lower-cost models. The point isn't one brand — it's learning to choose: you build an activity inventory that maps each QA task to the cheapest model that can actually do it, which is the part that makes a manager trust you with the bill.
Do I leave with a real, résumé-ready framework?
Yes. You build and operate a Playwright + BDD agentic framework (planner / generator / healer over MCP) against a real app, you build your own MCP server under it, and you get the vocabulary to defend every part of it under interview pressure.
What do I actually keep afterwards?
The framework repo and prompt templates, 386 interview questions with answers traced back to real class moments, a 74-topic study guide and glossary, a mock-interview platform to rehearse out loud, your own account on ZincTM and its MCP server, and every session recording. All of it permanently, at no extra cost.
Does it stop after four weeks?
No. The Wednesday interview-prep sessions with Kuzzat keep running after the cohort ends, there's a two-day bootcamp on building the framework from scratch, and ad-hoc sessions whenever the tooling shifts. Kuzzat also reads résumés personally, and we check in at three months to hear where you landed.