AWS Builder Center
Think You Can Model It? Introducing the DynamoDB Football Data Modeling Challenge Series.

Think You Can Model It? Introducing the DynamoDB Football Data Modeling Challenge Series.

A four-part challenge series that puts your DynamoDB design skills to the test with real-world scenarios, expert judging, and prizes for those who prove they can build at scale.

DynamoDB Solutions Architect
It's one thing to read about single-table design, write-sharding, and conditional writes. It's another thing entirely to sit down with a blank canvas and design a data model that actually works at 10 million concurrent users with millisecond latency requirements.
That's exactly what we're asking you to do.
Today we're launching the DynamoDB Football Data Modeling Challenge — a series of four progressively complex design challenges hosted on AWS Builder Center each with an opportunity to win prizes! Each challenge drops a real-world scenario in your lap and asks you to produce a production-grade DynamoDB data model, complete with a NoSQL Workbench export, a design document describing your decisions, and an access pattern matrix. Show us the schema, explain the trade-offs, and prove it scales.

The Setting: A Global Sporting Event

All four challenges share a common narrative: a massive global sporting event with millions of passionate fans. From the moment tickets go on sale to the final leaderboard update, every system in the event's digital platform is backed by DynamoDB — and each challenge asks you to design a critical piece of that platform.
The scenarios aren't academic exercises. They're drawn from the real scaling problems that teams face when building systems for events with millions of concurrent users, spiky traffic patterns, and zero margin for failure.

The Four Challenges

 Challenge 1: The Virtual Waiting Room 

July 3–10, 2026
One million fans hit the system simultaneously. Before anyone can buy a ticket, you need to absorb the stampede, fairly assign everyone a place in line, and progressively admit them to the purchasing flow without overwhelming downstream systems.
You'll design a DynamoDB-powered waiting room that handles extreme write bursts, assigns verifiable queue positions that are resistant to gaming, and promotes fans in controlled batches based on available purchasing capacity. This is high-throughput ingestion meets fairness — a deceptively hard combination.



Challenge 2: The Fair Seat Purchase

July 17–31, 2026
100,000 seats. Hundreds of fans clicking "buy" at the same instant. Every seat must be sold exactly once.
Once fans clear the waiting room, they enter a live seat map. Your job: design the inventory control and distributed locking system that guarantees no seat is ever double-sold. You'll model the full lifecycle — browse → select → hold → pay → confirm — using and make it bulletproof under extreme concurrency.



Challenge 3: The Stadium Concessions Express

July 31–August 7, 2026
Halftime. A hundred thousand fans stand up at once. They all want nachos. Now.
Design an order fulfillment system that lets fans order from their seats, tracks which orders are pending, assigns orders to the nearest concession stand, and surfaces the longest-waiting orders first. The twist: DynamoDB doesn't natively support global sorting, so you'll need to get creative with your index design to serve time-ordered operational queries at scale.



Challenge 4: The Infinite Leaderboard

August 14–August 21, 2026
A billion fans. Unlimited creativity. Leaderboards that update faster than the crowd can roar.
Build a leaderboard platform that supports an unbounded number of named leaderboards, each handling over 1,000 score updates per second. You'll tackle one of the hardest problems in DynamoDB data modeling: high-write-throughput ranking systems where write distribution and read efficiency are in direct tension.



What You'll Submit

Each challenge requires three deliverables:
  1. A  NoSQL Workbench  data model — your complete table design exported as a  .json  file, including all tables, GSIs, LSIs, key schemas, and sample data.
  1. A design document — a written explanation of why you made each decision, not just what you built. This is your opportunity to describe trade-offs you made.
  1. An access pattern matrix — every access pattern mapped to the table/index, key condition, and filter expression that serves it.
You'll push these to a public GitHub repository, complete a brief submission survey on Builder Center, and post a comment about something you've learned.

How It's Judged

Submissions are evaluated across four categories — all Pass/Fail:
Category
What We're Looking For
Completeness
Did you deliver everything the challenge asked for?
Data Model Correctness
Does the schema actually work for the stated access patterns?
Scalability & Cost
Will this design hold up at the scale stated in the challenge?
Design Rationale
Do you understand and articulate the trade-offs?
A Pass in all four categories means your design is sound. Judges include both AI tools and AWS subject matter experts.

The Prizes

The first 50 qualifying entries per challenge will receive an AWS-branded jacket. Prizes are awarded in submission order, so there's an incentive to submit early — but not at the expense of quality. A submission must pass all four judging criteria to be eligible.

Ready to Start?

Challenge 1 opens July 3, 2026. Head to  AWS Builder Center  to register and get the full challenge brief. Download  NoSQL Workbench for Amazon DynamoDB  if you haven't already — you'll need it.
Each challenge runs for one week. You can participate in any or all four, but can only win prizes once. One entry per person per challenge.


The DynamoDB Data Modeling Challenge is open to participants in select countries who are 18+ with an AWS Builder Center profile. See the full  Terms and Conditions  on the challenge page for eligibility details and restrictions.
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article