Browse applications built on Generative Agents technology. Explore PoC and MVP applications created by our community and discover innovative use cases for Generative Agents technology.
A robust, Dockerized AI agent that automatically extracts visual frames from videos and leverages the Fireworks AI API to generate dynamic captions across multiple stylistic tones (formal, sarcastic, tech-humor, and everyday humor).
IOLITE is an AI-powered workout planner that creates personalised, periodized training programmes tailored to your sport, goals, and performance level.
CredIQ is a real-time credit card spending tracker with a live budget bar, AI chatbot, photo-based card analyzer, and offer recommendations — built on AMD infrastructure via Fireworks AI and Firebase.
CashFlow AI is an agentic AI platform that automates debt collection for businesses through intelligent reminders, payment tracking, accounting integrations, and human-in-the-loop workflows to maximize recoveries and reduce manual effort.
An AI-powered assistant that analyzes GitHub repositories and generates project summaries, marketing copy, and growth strategies for founders and developers
An AI agent that watches a short video and captions it in four distinct voices: formal, sarcastic, tech-humor, and everyday-humor
An offline-first AI agent that keeps small Ethiopian businesses recording sales during internet or mobile payment outages, using hybrid AI routing to cut costs by 86% while never letting the core workflow break.
QuadFrame Vision is an AI-powered video understanding engine that automatically extracts key frames from videos and generates high-quality captions in multiple styles using Gemma 4 on AMD ROCm, optimized for fast, containerized deployment.
BidCraft AI uses artificial intelligence to analyze tender documents, extract key requirements, generate compliance checklists, draft proposal sections, and help businesses prepare stronger, more accurate government and private sector bids.
Findable is an AEO (Answer Engine Optimization) audit tool — it checks how accurately and completely AI models (like ChatGPT/Gemma) actually know and describe a real person, then tells them how to fix the gaps.
Turn one sentence about a business idea into a complete startup kit — name, landing page, brand palette, and social captions — generated by four AI agents running on an AMD Developer Cloud GPU.
An aesthetic cognitive captioning workspace and real-time audio telemetry platform powered by Fireworks AI and AMD Cloud computing. Featuring multi-style subtitle tuning, instant Firestore sync, and dynamic interactive Recharts sentiment tracking.
MarketScout AI is an autonomous multi-agent platform that transforms startup ideas into market intelligence. Powered by Fireworks AI on AMD Cloud, it researches competitors, patents, funding, trends, and strategy to validate ideas in minutes.
MonadLabs converts all data into data specialized for agents, integrates directly in company workflows, and reduces hallucination and token spending up to 85%. As API Bills surge into billions, it is the only way for enterprises to use AI efficiently.
A zero-external-token routing agent combining deterministic solvers, verified local Qwen inference, and AMD ROCm acceleration.
Agentic Microcontroller Tester Agentic Microcontroller Tester
swarmtrace is a ai agent monitring platform build for to give you complete control of your ai agents saas
YOUSUN Axion turns messy AI workloads into AMD-ready deployments by scanning code, detecting CUDA/ROCm issues, suggesting fixes, benchmarking improvements, and creating a professional readiness report.
KidoDev is an AI coding platform inspired by Scratch that helps kids develop coding, problem-solving, and critical thinking skills. Powered by Gemma-4 26B-IT, AMD Cloud, AMD Fireworks AI, GEMMA-4 AI Tutor, Co-Mentor.
AI arbitration infrastructure for escrow. Autonomous agents read contracts, verify trade documents, and route every deal to auto-settlement, pending evidence, or human arbitration — a C++ engine holds the money. Agents route. Humans rule.
Pronounce AI helps broadcasters and journalists confidently pronounce unfamiliar names, places, and organizations. Powered by AI, it delivers phonetic guides, IPA, presenter tips, and intelligent caching for faster, more reliable newsroom preparation.
An AI agent that classifies each task and routes it to the optimal Fireworks model — using the most capable model only when accuracy demands it, and the cheapest model everywhere else to cut token usage.
An autonomous AI agent that analyzes your resume against a job, finds skill gaps, builds a personalized roadmap, self-checks its own work, and preps you for interviews
An LLM proposes, an exact solver decides, an independent verifier certifies — business decisions you can prove feasible and optimal, not just plausible answers from a chatbot.
Clash Chronicles transforms any Clash Royale deck screenshot into a cinematic adventure Using advanced computer vision and Generative AI, extracts cards in your deck and instantly generates a unique immersive three-act story starring your chosen squad.
AcademicForge is an AI-powered academic writing assistant that helps students and researchers plan, draft, refine, and organize high-quality scholarly work faster.
Dockerized async video-captioning agent: ffmpeg keyframes->dual Gemini vision models describe scenes->rotating pool (Groq Llama, Mistral, Gemini Flash) writes 4 caption styles.Queue-based concurrency,retry/backoff, free-tier limits,10-min runtime budget.
ForgeMind AI is an AI-powered software architecture assistant that transforms project ideas into complete technical blueprints, including system architecture, security recommendations, development roadmaps, and implementation guidance and etc.
RadeonShift AI is an AI-assisted CUDA-to-HIP migration tool that translates legacy NVIDIA code to AMD ROCm. It audits architecture risks and provides live MI300X telemetry to drastically accelerate enterprise hardware transitions.
ByteRoute is a smart AI routing agent that minimizes Fireworks API token usage by intercepting simple tasks locally using regex and heuristics, while routing complex tasks to GLM 5.2 with category-specific system prompts for maximum accuracy.