Devpost
Participate in our public hackathons
Devpost for Teams
Access your company's private hackathons
Grow your developer ecosystem and promote your platform
Drive innovation, collaboration, and retention within your organization
By use case
Blog
Insights into hackathon planning and participation
Customer stories
Inspiration from peers and other industry leaders
Planning guides
Best practices for planning online and in-person hackathons
Webinars & events
Upcoming events and on-demand recordings
Help desk
Common questions and support documentation
Sort by:
We reverse-engineer the slot machine’s hidden rules, then use exact Bayesian inference to turn each machine’s payout history into calibrated odds for its next spin.
Luck, location or a leak? Our dashboard splits the casino's $541K slot gap into game math, location, floor time and luck. It finds no broken machines and backs an $8,700 test plan over replacement.
Don’t replace the 10 lowest slots. Fix what’s really wrong.
Interactive dashboard for a 300-machine casino: why the bottom ten earn least, and how to spend a $100,000 July budget. Keep the ten, inspect one machine, run one tournament weekend.
Pip turns rubber ducks into screen-free learning buddies. Students get personalized support while teachers manage every duck through the Pond! app.
A poker bot that simulates the odds, learns its opponent, and lets a neural network second-guess its moves.
We find each slot machine’s hidden lag rules automatically, then predict payouts with LightGBM + transformers. New rules mid-event? Re-found and retrained in 8 min.
Smart indoor navigation that predicts event-based congestion and recommends less-crowded routes using location-specific data.
a poker bot that taught itself short deck 5 card PLO by playing hands against itself with CFR, then adapts to how each opponent plays
The house always win. Casino Manager is why.
A multi-agent control tower that turns messy disruption emails into exact, audited fixes: Gemini reads, a knowledge graph finds what's affected, and rule-bound agents act with zero failed calls.
pokebot for heads up challenge
Reading the house edge: Bayesian forecasts of hidden slot-machine rules that adapt to every new round.
Turning casino operations into actionable insights through analytics, machine performance investigation, and forecasting our own recommendations to management.
We rebuilt our casino-floor analysis as a governed Databricks lakehouse: checksummed Delta layers, a 12-check quality gate, MLflow, grounded Llama explanations and Genie, all matching our Python.
An adaptive heads-up poker bot for Short Deck PLO5, combining Monte Carlo equity simulation, opponent behavior modeling, and expected-value-based decisions under strict time limits.
AI learns to play block-stacking game (allegedly) (sans AI)
Our project is a Python-based AI poker bot designed to compete against other automated players in heads-up poker.
Approve one budget for AI product cards and pay only for the cards that pass your checks.
When something at home is stolen or ruined, Paper Trail searches your NAS for receipts, policies, and photos. Then it builds a claim folder where every item links to its proof. No proof? It says so.
Because Security is your Beeswax
Your data is out there. Find it, trace it, take it down.
A Poker Bot for the Heads Up Challenge coded using a GTO Monte Carlo Mix strategy, programed with the help of Cursor with Claude as support for the Build Your Hackathon A-Team with SpaceXSI Challenge.
