A six-week live cohort by AWS Fundamentals
Shipping on AWS With Confidence
Stop second-guessing your architecture. Ship a professional-grade fullstack app on AWS and defend every decision behind it.
Taught by Sandro and Tobi, live.
Founding cohort: $797 $2,497. The founding price happens once.
Trusted by developers at
From toy tutorials to building like the industry.
You can already build fullstack apps. The hard part is not deploying to AWS. It's getting from it works to it's resilient, observable, secure, and cost-efficient in production, and knowing that before every release instead of after.
AWS has 200+ services. Almost every app out there runs on the same ~20. Learn them once with us. Reuse the pattern everywhere.
The problem
You know how to build fullstack. You never feel like you're using AWS right.
And honestly? You're probably not.
Sound familiar?
...and many more.
Real horror stories
Async-service sprawl
SQS here, SNS there, EventBridge somewhere else. Fine for one feature. A painful refactor once the whole system went async.
Single-account blast radius
Five dev sandboxes, staging, and prod in one account. One load test poison-pilled everything. Prod went down. Customers noticed.
The $1,300 empty S3 bucket
One empty bucket with a guessable name. Someone else's misconfigured backup tool hammered it with almost 100 million requests. The bill: $1,300. In one day.
Read the real story →
The terabytes that never came
Every project planned to be the next Facebook. Single-table DynamoDB, designed for millions. Then the team mirrored everything into RDS just to sort a table. The scale never arrived. The complexity stayed.
Don't be that guy.
Believe us, we've been there. AWS doesn't have to be hard. It's the same pattern every time: learn it once, apply it everywhere.
Why a cohort
This is not another course.
We've launched books and courses before. They all have one thing in common: most people never finish them. We saw it with AWS Fundamentals. We saw it with the CloudWatch Book. So this time we do it together, 15 people at most, small and close. That means:
The workshop call
You watch a short pre-work video, then we make the week's architecture decision live, together, in the workshop call.
The clinic call
Later that week you bring the decision you made and we tear it apart, in the friendliest way. No other course does this part.
Homework. Yes, really.
One decision record a week, submitted as a pull request. Sandro or Tobi write and sign every review by hand.
Community and alumni
A private Discord with your cohort and us. Ask anything, discuss your decisions, and stay after the cohort ends.
How you will work
You decide what you'll use. We help you.
We don't just tell you to use a service. We teach you to decide for one. That's what architecture decision records are for. What you chose. What you rejected. Why. And what would change your mind. AI is a multiplier here. If you can judge its output, it makes you 10x faster. If you can't, it multiplies your mistakes at the same speed.
“An agent is a mid-level developer on steroids. Super confident, doesn't know what it doesn't know, and ships a thousand lines that should have been ten.”
Steering it is the skill. We asked 36 developers on our list: 44% can't tell whether what AI builds for them on AWS is right. You submit one decision record every week as a pull request, and Sandro or Tobi write and sign the review by hand. Here's a real one.
# ADR-002, primary datastore for the app
## Decision
Amazon RDS for PostgreSQL, with Drizzle as the ORM. Boring on purpose.
## Rejected
DynamoDB. Optimizing for scale we don't have, at the cost of velocity we need now.
Aurora DSQL. Compatible is not Postgres. No foreign keys, several SQL features missing, ORMs don't fully support it yet.
## What would change my mind
A hard multi-region active-active requirement, or spiky unpredictable load where serverless-first matters more than relational developer experience.▋
An architecture decision at every layer.
You learn the decisions behind every part of a fullstack application. We don't skip. We go from the frontend, to the backend, to async, all the way to observability and running it in production.
Where everything starts
Your landing zone, the version you actually need
Account structure, AWS Organizations, and IAM Identity Center for SSO. We run plain Organizations instead of Control Tower, and we tell you when that flips. Then profiles, CDK bootstrap, budget alerts, and a hello-world shipped through CI/CD. How you structure your accounts is an architecture decision too, so we start right there. You leave with your own sandbox to break things in, and a preview environment per pull request. The setup most teams never get right.
API and compute
When do you reach for Lambda, when for a container, and is there even a place for Kubernetes? We make the call with you.
Frontend and delivery
What's server side, what's client side, and when to use which. The frontend got unnecessarily complex. We help you decide, and show you how to deliver it on the edge.
Database
The big one. NoSQL or SQL, which database, and is Postgres-compatible really the same as Postgres? We help you decide.
Async and events
Event-driven architecture is a great tool, but start it wrong and you end up with SQS, SNS, EventBridge, and Kinesis all in one place. We show you an architecture that holds.
Observability, security, and production
CloudWatch done right: logs, metrics, alarms. When production breaks, you always suspect your own code first. Good observability tells you where to actually look. Then the guardrails we deliberately left out at the start, now that you have something worth guarding: SCPs, Security Hub, GuardDuty, and org-wide policy. Plus quotas, cost, and what "ready for real traffic" actually means. The part most courses skip.
One reference app
Not a toy. One real application, built properly, together.
Every decision goes into one real application we build together, week by week. A public status page, an event-driven incident flow through EventBridge, alerting that tells you before your customers do, and a multi-account setup underneath everything in this system.
See the app we build →The reference app
This is what we build together.
One real product across all six weeks: an uptime monitor with a public status page. The screenshot below is from the actual app, not a mockup. And it's fully integrated with AWS:
The best part: you choose and defend the services you use. We have a recommended architecture, but AWS gives you options. We help you decide which one to pick, and when.
The check engine writes results from three AWS regions into Postgres. Uptime bars, latency percentiles, real data.
Ready to finally build on AWS the right way?
15 seats in the founding cohort. Join the waitlist and lock in the $797 founding price.
Who this is for
This is for you if
- You ship fullstack apps, but the moment it hits production the AWS side feels duct-taped.
- You know the services. Choosing between them still feels like a coin flip.
- Your app works in production. You're not sure it's resilient, observable, or cheap.
- You're the one people ask about AWS, and you still widen an IAM policy until it finally works.
- You passed the certs and still guess on the real decisions.
- You want someone senior to look at your actual architecture, not a toy example.
- You lean on AI for infrastructure and can't tell whether it's right.
This is not for you if
- You want cert prep.
- You don't want to build software.
- You want someone to make the decisions for you.
- You don't want to use AWS.
What you should already have
You can program and read code in one language. You've shipped something before, even if it was just a website. And you're curious. That's it. You don't need to be an AWS expert. That's our job.
Your guides, two lenses
Nobody else can honestly say “containers here, serverless there, depending on your context” and back it up from having lived both sides.
Sandro
The startup lens
I'm a freelance cloud engineer. I've spent years building serverless on AWS for startups, APIs that handle over 100 million requests a month, event-driven systems doing over a million events a month. These days I build landing zones, internal AI platforms, and fullstack apps that stay cheap and boring on purpose.
Tobi
The enterprise lens
I come from the enterprise platform side. I've migrated huge workloads from on-premise to AWS and built IAM used by over 30,000 monthly active users. Today I run one of the largest AWS accounts out there, big Kubernetes clusters where Kubernetes actually earns its place, centralizing DevOps so feature teams ship fast again.
Between us we've worked on over 100 AWS accounts, set up multi-account organizations, shipped many indie projects we loved, and taught more than 50,000 developers at AWS Fundamentals.
See all our talks →
What you get.
Everything in the founding cohort.
The six-week live cohort
Twelve live sessions. We finish this one together, with a community to ask questions and us to answer them.
Week 0, before we start
A setup walkthrough so everyone arrives ready: AWS account, profiles, and the AI toolchain we build with. Plus our AWS Fundamentals book, free.
Ship one fullstack app together
A real application on AWS, built week by week, that you take with you.
Six defended decision records
One for every major architecture call, reviewed with you. You walk out with six ADRs you can show your team.
Optional one-on-ones
Support, mentoring, unblocking, whatever you need. Real time with us.
A private community, forever
One Discord you only reach through this course. You keep that access for good. Ask anything, get help.
Lifetime access to everything
All the recordings, every decision record, and the code. Yours for good.
An official completion certificate
It shows what you learned and how you learned it.
A guarantee in three layers
Week one, no questions
Not for you after the first week? Full refund, no questions asked.
Did the work, still unsure?
Finished all six records and the weekly builds and still not confident? A free one-to-one architecture review with one of us.
Still not satisfied?
Join the next cohort for free, or take a full refund.
Don't take our word for it.
Hear it from fellow developers, AWS leaders, and more.
“Focused on AWS usage in the real world. Goes deeper than simply providing the information needed to obtain certifications. The per-service infographics also look amazing.”
“This is awesome, Tobias!”
“Amazing!”
“I am excited about this one. We get a lot of web dev and JS around here, and we are missing more cloud, infrastructure, and engineering fundamentals.”
“Jezzzz this is amazing and I will love to translate it.”
“Keep an eye on this release if you are interested in learning about AWS fundamentals.”
Founding cohort
15 seats. 10 minimum. The peer experience breaks above 15 on a live call. The founding price happens once and never comes back.
Early believers get in before the testimonials exist and shape the curriculum with us. Every later cohort pays more.
Waitlist only for now. We'll email you when founding seats open.
Reserve your founding seat.
15 seats. Waitlist members hear first when the cart opens.
Not ready yet? Start with the newsletter.