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
An SLA-aware middleware that dynamically picks the strongest LLM guardrail tier that fits a real-time latency budget turning security strictness into a tunable and measured tradeoff on Arm.
AI Financial Operating System for the next generation of personal finance.
One dependency-free C core turns heartbeats into a readiness state — Arm64 server, WebAssembly, or Cortex-M, unchanged. 19× faster, 5× smaller, proven identical to scikit-learn.
Native Arm64 CI that quantizes an ONNX intent classifier to INT8 and publishes reproducible accuracy, size, latency, throughput, and memory trade-offs.
Evidence-first confidence-gated AI inference for Arm64 CPU services.
Evidence-first micro-batching benchmarks for Arm64 CPU AI inference.
Evidence-first FP32 vs INT8 benchmarking for on-device AI on Arm64.
Adaptive on-device AI for Arm phones that routes between Tiny INT8 and Large FP16 models using confidence and live device conditions—3.69× faster, 93.5% smaller, fully offline.
Other people aren't even trying, well I am.
An offline-first adaptive study coach with a 324-byte INT8 decision policy for Arm-powered devices.
Benchmarking and optimizing LLM inference (llama.cpp) on Arm64 vs x86 cloud infrastructure, using free GitHub Actions runners fully reproducible, no cloud account required.
PocketIDE turns an ARM-powered phone into a private offline IDE where a local AI agent writes, runs, tests, debugs, and repairs real code without a server.
Educator AI — Advanced Autonomous Career Mentor & Global Intelligence Gateway
I built AIGVDetPlus, an optimized, research-backed deep learning pipeline designed to detect deepfake videos by analyzing spatial and temporal anomalies in manipulated media to protect digital trust.
"Aristotle: a browser-based algebra tutor that runs entirely on-device via ONNX — no server, no latency, no data leaving the student's device. Socratic guidance that works offline
8.80× faster native Arm64 agents with 85% held-out workflow success.
What if you summoned a city into existence by typing? AI agents design it live, then flood-test it—so your design isn’t judged by slides, but by survival metrics.
An Arm-powered physical AI robot using Python for local voice recognition and brain logic, connected via Wi-Fi to an ESP32 microcontroller for real-time body, servo, and audio control.
ArmTune: reproducible build, runtime, and system tuning for AI inference on Arm.
A from-scratch LiDAR 3D object detector, quantized to int8 with QAT and deployed on real Arm silicon. About 4x smaller, 4x faster on one Arm core, keeping 92% of accuracy.
OptiMind AI optimizes AI models for faster inference, lower power consumption, and better performance on Arm-powered edge devices, enabling scalable, efficient, and sustainable AI applications.
Quality-gated GGUF optimization for reliable AI inference on native Arm64.
A platform that helps small business Nigeria get online find customer and grow faster
Give me an Arm phone and a GGUF. ENTITY reads the silicon, tunes itself to it and proves which decision earned the speedup.
241 – 264 of 333
