Browse applications built on modern technologies. Explore PoC and MVP applications created by our community and discover innovative use cases for modern technologies.
DocuPilot AI transforms PDFs, scanned documents, and invoices into searchable knowledge, evidence-backed answers, financial insights, and explainable classifications through a secure, human-in-the-loop document intelligence platform.
SnapSell turns narrated inventory videos into reviewed resale listings, researches evidence-backed asking prices, prepares marketplace listing kits, and helps sellers manage buyer replies, pickups and recorded sales with AMD-powered AI.
A four-seat Claude Code factory in Band Desktop (coordinator, implementer, tester, reviewer) that built all four stages of the Pocketful wallet and payments spec autonomously: one task message per stage, 114 minutes, zero human steering.
Three AI agents plan, execute, and audit payment transfers in a BAND room. A C ledger and FastAPI service enforce atomic transfers, idempotency, and exact integer-cent accounting, supported by 76 passing automated tests.
Five BAND Desktop AI seats built a wallet service; all four stages accepted by an independent gate. No seat accepts its own work: 7 real defects caught by review, cost measured per seat.
Three AI agents, one rule: whoever builds never tests. Our Band-room factory shipped a four-stage payments service with two human messages, while its reviewer, the only seat that tests, rejected the work six times until it held up.
GhostProcess is an autonomous 4-seat AI dark factory on BAND Desktop. It uses a blind red-team agent to attack and certify a clean-room double-entry ledger backend and responsive UI with zero human intervention and zero API cost.
Four autonomous agents built all four Pocketful stages from one human message. They rejected two green candidates — one crashed after 4,761 reads while every public check passed — and accepted only evidence they had reproduced.
SheetShift uses IBM Bob to turn an insurance rating spreadsheet into owned, tested Python, then verifies it cell by cell: 430,000 cells compared, 0 unexplained, every anomaly decided by a person.
StockMind is an AI-powered stock analysis platform that combines LSTM forecasting, technical indicators, market sentiment, risk analysis, and backtesting to help users understand stock trends through an interactive fintech dashboard.
Autonomous agent platform running deep ledger data analysis and AI Q&A to detect margin anomalies, audit cash flows, and generate executive growth strategies.
Nigeria bills + USD→NGN remittance assistant: quiet when your naira is safe, a ping when bills or FX need you. Built with IBM Bob IDE.
Nigeria remittance timing agent with Intel OpenVINO on-device intent routing — Intel Online track for AI Infra Summit.
Voice-first fleet management platform powered by AssemblyAI and Gemini that instantly converts noisy cabin driver speech into structured vehicle expenses, fuel logs, and telemetry metrics.
Monevo is a voice-first WhatsApp AI financial assistant powered by AssemblyAI. Record expenses, log income, request analytics, and generate digital receipts simply by speaking natural voice notes right inside WhatsApp with zero apps to install.
TRACE is a voice-native AI investigation agent that finds why business numbers don’t add up by tracing fragmented records, connecting evidence, testing hypotheses, and explaining what happened.
A runtime governance layer for autonomous AI agents that evaluates intent, permissions, risk, budget and evidence before actions execute, using AMD ROCm and AMD Developer Cloud to power transparent, auditable agentic workflows.
AI proposes. Evidence decides. Humans authorize. A governed AI options trading agent with 12 deterministic risk gates, human consent, and SHA-256 audit trail on Alpaca paper trading.
Multi-agent quant trading engine powered by Alpaca & LangGraph. Features 5-agent parallel debate, dual-track equity & options execution, real-time market regime adaptation, deterministic RiskGuard, and 24/7 Cron Sentinel position governance.
Petra is an autonomous options alpha trading agent originally built as a submission for the Alpaca AI Trading Hackathon. Here is a brief overview of how it works:Core Function: It automatically sells defined-risk credit spreads.
An autonomous, multi-agent AI options trading engine. Aegis Alpha uses an LLM debate protocol to prevent hallucinations and enforces strict quantitative risk gates to safely execute complex multi-leg trades via the Alpaca API.
Alpaca AI is a multi-agent options trading desk that debates opportunities, applies risk controls, and autonomously executes portfolio-aware paper trades.
Drawdown Guard is an autonomous AI agent that keeps the portfolio within a client-defined downside limit through an options overlay strategy
Autonomous multi-agent options trading system on Alpaca combining real-time regime detection, Black-Scholes Greeks, Kelly Criterion sizing, and a 6-gate risk engine to generate consistent alpha on a $100K paper portfolio.
An autonomous options trading agent that reasons over real SPY indicators and live news with an LLM, sizes trades by confidence, auto-manages exits, and trades on Alpaca's paper API and CLI — validated with real overnight runs and backtesting.
Two near-identical option trades: one costs $1 to get in and out of, the other $306. Options have no legally required execution receipt, so I measured 65 real contracts on live markets and built one, plus an agent that trades under that discipline.
An autonomous put-credit-spread desk on Alpaca paper. Seven deterministic gates run before any order exists; the single model call classifies news and can only veto a candidate, never create one. Most cycles end in NO_TRADE. Every one is journalled.
An autonomous options paper-trading desk that debates its trades: a quant funnel finds candidates, GPT analysts propose, a Claude Red Team attacks, and a deterministic Risk Constitution decides — then measures whether each layer added value.
Signal Hunter is an autonomous AI agent that detects unusual options flow, reasons about catalysts and IV with an LLM, and trades within hard, code enforced risk limits via Alpaca's Trading API and MCP server.
An autonomous options trading agent for SPY and QQQ. A council agent proposes trades and a critic agent reviews them, but a deterministic chain has the final say on what actually reaches the broker. Ran unattended for four sessions, finishing up 3.26%.