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:
World's fastest full-cycle processing at 3.82ns on standard ARM, zero overhead, self-reset in one shot
A lightweight AI runtime for ARM Cortex-M that uses dynamic memory allocation and reusable memory pools to reduce SRAM usage and enable efficient on-device AI inference.
EdgeTuner searches the ideal way to run your model on a Raspberry Pi 5. Quantization, graph optimization, memory arena, and threads are all measured and optimized for Raspberry Pi.
Offline AI accessibility app for blind, deaf & mute users — object detection, TTS, speech-to-text & sign language recognition, running entirely on-device via ARM optimization.
Most optimize the model. We optimized the wire: a binary, NEON-aligned format for MCP tool calls. ~12× more agent-turns per Neoverse core, 44% fewer bytes, confirmed on Arm Performix counters.
Self-calibrating llama.cpp serving for Arm CPUs. Probes memory bandwidth, sizes the deployment from the GGUF header, auto-finds the latency knee under open-loop load.
EdgeSplit runs LLM prefill on a laptop and decode on a phone, turning everyday hardware into a distributed inference system.
AEOLUS simulates a space habitat's air system to spot faults like a broken fan early, before the air gets unsafe. Every run is repeatable, so the results can always be trusted and checked.
An AI-assisted port of Wolfenstein 3D to low-power touch devices, focused on SDL optimization, performance tuning, and a clear human-AI development workflow.
Building intelligent, hardware-assisted ransomware defense for ARM-based embedded systems through real-time detection and automated protection.
An offline real-time walking-hazard alert system for visually impaired users on Raspberry Pi 4, powered by Arm INT8 quantization & XNNPACK acceleration for 3.76x speedup & 28.9ms latency.
Edge AI-powered real-time threat detection on Renesas EK-RA8P1 using Arm Ethos-U55, detecting guns and knives with optimized on-device inference and instant alerts.
A router-driven Qwen2.5 cascade on AWS Graviton4 that beats a self-hosted baseline by 20x and a hosted API by 4x — built with llama.cpp, KleidiAI kernels, and speculative decoding for Neoverse-V2
Full-Stack Physical AI Optimization on RA8P1: INT8, Vela, Ethos-U55 and RTOS Parallelism
EcoVision AI is an Arm-optimized edge AI system that uses a camera to identify and classify waste (plastic, paper, metal, glass, organic, and e-waste) in real time
An RTOS-based Edge AI system that runs an autoencoder directly on an STM32 to detect abnormal water quality from pH, TDS, and turbidity in real time.
Optimize MobileNetV2 for faster, smaller inference across x86 and Arm, with automated benchmarking, cross-platform comparison, and deployment recommendations.
UniEdge: Adaptive LLMs on Arm. Optimizing Gemma-3 via KleidiAI/NEON & local SQLite memory for 17% faster inference and offline continuous learning on resource-constrained edge devices
Detect the silicon, benchmark on-device, apply the winning config — a phone that gets faster, not just measured.
Optimizing YOLOv8 for faster and lighter AI inference on ARM-based edge hardware using INT8 quantization, ONNX Runtime, and multi-threaded CPU execution.
On-device relational AI infrastructure—enabling everyone to have an AI companion that truly understands them, is always by their side, and is theirs to fully own.
Four users of one AI agent cost 10.9× time-to-first-token on Arm Neoverse N2. One flag fixes it. We swept 6 prompt shapes to find exactly when it matters — and when the flag does nothing at all.
Black Dragon Runtime: the world's first Distributed Industrial Nervous System, optimized for Arm
LocalScribe is a privacy-first, on-device AI application that transforms voice recordings into accurate transcripts, concise summaries, and actionable notes
