Browse applications built on OpenAI Custom GPTs technology. Explore PoC and MVP applications created by our community and discover innovative use cases for OpenAI Custom GPTs technology.
Aura is a sovereign, neuro-symbolic AI operating system that turns natural-language intent into transparent, governed action through local parsing, agent collaboration, bounded experimentation, verifiable learning, and human-controlled decision gates.
Clinical decision support for elderly polypharmacy. A deterministic engine 102 drug interactions,61 AGS Beers rules, 58 STOPP/START v3 criteria runs in ~3ms on every request, gated on age and eGFR, escalating to a cloud LLM only when the case warrants it.
BoundaryForge AI evaluates regulated financial AI before deployment using Gemma 4 adversarial prompts, OpenAI/Claude targets, independent judges, and AMD ROCm validation to produce risk findings, certificates, and audit evidence.
Maya is a real-time, voice-first AI speaking tutor — you talk, she listens and replies out loud, then instantly shows what you said, what you meant, and how a native speaker would actually say it.
A CLI tool that detects CUDA-specific Python code that breaks or degrades on AMD ROCm, flagging each issue by severity with a sourced explanation and fix guidance - filling the gap HIPIFY leaves at the application layer.
Esillio is a local-first biological memory layer. We leverage AMD Instinct™ GPUs and ROCm to distill Gemma 4 into a micro-artifact that runs fully offline on consumer CPUs, creating a highly defensible, patient-owned moat.
Veedra watches a short video clip and writes captions in four styles: formal, sarcastic, humorous-tech, and humorous-non-tech. It grounds every caption in what's actually on screen using a two-step vision pass, so the jokes stay true to the footage.
Building Human-like AI companions with human-inspired cognition and self-learning capabilities that grow, adapt, and evolve through experience.
Livora AI is an AI health companion that helps users understand, monitor, and improve liver health by combining medical reports, ultrasounds, biomarkers, and lifestyle data into personalized risk scores, progress tracking, and preventive guidance.
A validator-gated hybrid AI agent for AMD Hackathon Track 1 that combines deterministic local solving with Fireworks escalation to maximize unseen-prompt accuracy while keeping token usage efficient and submission-safe.
Where edge computing meets aerodynamics. We utilize the raw power of AMD processors to render real-time digital twins of localized wind conditions. Our submission, Alpha Vario, translates massive environmental datasets into immediate, actionable help
Paste a startup's URL and DealScope returns a source-traceable deal screen in 30 seconds — every claim links to its public source — powered by intelligent model routing on AMD-hosted GPUs.
An AI-powered escrow marketplace solving cash-on-delivery risk for Africa's informal sellers; autonomous agents (ZEMA 360) run fulfillment and pricing on Qwen 3.7 Plus, served on AMD Instinct GPUs via Fireworks AI, live in production today.
An AI Systems Research Platform focused on maximizing Capability Density on commodity hardware.
ClipForger-AMD turns short videos into accurate burned-in captions plus four styled summaries: formal, sarcastic, humorous-tech, and humorous-non-tech, using Fireworks/Gemma-first routing with fallback reliability.
Quorum is a Band-powered governed analytics platform where specialist AI agents collaborate to answer business questions, generate SQL, validate compliance, and deliver auditable insights with complete decision traceability.
Autonomous compliance intelligence — from regulation drop to board decision in minutes.
A multi-agent corporate relocation orchestrator that automates travel planning and budget policy auditing using Band SDK, AI/ML API, and Featherless AI.
Band agents that remember. Mimir-powered persistent memory gives Planner, Executor, and Reviewer agents shared context across sessions — they recall past decisions, conventions, and lessons so your multi-agent team gets smarter every run.
BrightSenani is a competitive intelligence command center for GTM teams, competitor tracking, pricing and positioning insights, launch and hiring signals, threat scoring, AI battlecards, prospect intelligence, and an executive advisor in one workspace.
AlphaLens is an autonomous earnings signal engine that scrapes major social sites of big firms and synthesize them into a single 0–100 score predicting earnings surprises. What quant hedge funds build manually in months, AlphaLens at does night.
An AI powered interactive puppet theater where users control expressive 3D characters using real time hand gestures and facial expressions to create cinematic live performances inside a dynamic virtual stage.
FeatureFlag is an autonomous AI copilot for continuous deployment that uses reinforcement learning, anomaly detection, and multi-agent simulation to automate feature rollouts, instant rollbacks, and enterprise-scale release governance.
Bob Truth Score audits closed Jira tickets against linked GitHub PRs and repo context, exposing whether work is truly done, partially complete, or only marked Done.
An agentic cursor that sees your screen live, understands your goal, and moves the cursor autonomously until the task is complete. Just like Copilot helps you code, AXON helps you operate your entire computer.
AI-powered sprint management dashboard that predicts and prevents sprint failures before they happen, featuring real-time risk analysis, team workload optimization, and intelligent recommendations for agile teams.
RepoMind : an AI-powered repository analysis platform that instantly explains any GitHub codebase with architecture insights, tech stack detection, onboarding guides, health metrics, and interactive AI chat—making unfamiliar projects easier to understand.
AI-powered smart energy monitoring system that tracks room and block-wise electricity usage, automates lights and fans using motion detection, and provides real-time analytics with optimized energy-saving recommendations.
MindShield by BrainSNN is a 3D cognitive firewall for enterprise AI, using Gemini analysis, Veea Lobster Trap, and self-evolving defenses to detect manipulation, block unsafe prompts, and fuse multimodal knowledge.
A platform where AI agents learn skills from certified AI teachers. Enroll, study structured courses, chat with an LLM teacher, and earn certificates. Works with Claude Code, OpenClaw, or any REST client.