Caroline Sithole@Matriarch4321

Petra is an autonomous options trading agent developed as a submission for the Alpaca AI Trading Hackathon. It operates on a $100k Alpaca paper trading account to automatically sell defined-risk credit spreads, which include put credit spreads, call credit spreads, and iron condors. The agent is built on a full-stack architecture utilizing FastAPI, MongoDB, React, and Claude Sonnet 4.6. Architecture & Risk ManagementThe core philosophy behind Petra is that options selling accounts typically blow up due to sizing and liquidity mistakes rather than bad directional calls. Therefore, the system strictly separates AI intelligence from financial risk: Signal Generation: Claude Sonnet 4.6 acts as the brain, determining the market regime, direction, confidence, and strategy choice. Strike & Size Engine: Deterministic code calculates the delta-targeted short strike, spread width, and contract count based on a strict risk budget. Execution & Management: The system routes multi-leg limit orders at a net credit. It continuously manages positions utilizing a 50% take-profit target and a 2x-credit stop loss. The Trading CycleThe agent operates autonomously on a strict schedule:An in-process scheduler triggers a cycle every 15 minutes while the market is open. It captures a snapshot of a defined universe of assets (such as SPY, QQQ, AAPL, and NVDA) and requests a trading verdict from Claude. The deterministic engine builds the spread and evaluates it against a 7-rule risk gate, where any failure immediately halts the order. Safety & Transparency Fail-Closed Design: Any LLM schema violation, data fetch failure, or risk gate rejection guarantees that no trade is executed. Full Audit Trail: Every prompt, verdict, risk check, and exit reason is saved in a MongoDB database and displayed on a live dashboard for complete transparency. Would you like to explore the specific parameters of Petra's 7-rule risk gate or dive deeper into how the dashboard visualizes the data?

CareGuard is a multi-agent AI system designed for workplace psychosocial case management. It helps HR teams efficiently handle sensitive employee wellbeing incidents by orchestrating six specialized AI agents that work together to: Triage and classify cases Assess psychosocial, legal, and retaliation risks Ensure compliance with duty of care, confidentiality, and EAP obligations Recommend appropriate care pathways and referrals Produce structured advisory memos and RACI action plans How it works CareGuard can operate in two modes: Band.ai SDK Mode (recommended): Creates a dedicated Band.ai chat room where the six Python agents (Triage Sentinel, Risk Analytics Engine, Policy Guard, Core Navigator, Compliance Review Director, and HR Advisory) deliberate in real time via @mentions. Local LLM Mode: Runs the full multi-agent workflow inline using models like Claude, Qwen, DeepSeek, Llama, or GPT-4o-mini. Built with React 19 + Tailwind (frontend), Express + TypeScript + MongoDB (backend), and deployed on Vercel, CareGuard streamlines complex HR case reviews while maintaining high standards of privacy, compliance, and ethical handling.

NexusIntel is an advanced, multi-track intelligence platform designed to deliver comprehensive, real-time insights on any target company. It leverages Bright Data’s scraping ecosystem and Claude AI to run three parallel data pipelines. Orchestrated via LangGraph, the system synthesizes vast amounts of live web data into a unified, actionable C-suite executive brief. GTM (Go-To-Market) Intelligence: Focuses on competitive positioning and buying signals. By scraping competitor pricing, analyzing live job postings (to infer strategic pivots), and monitoring SERP for vendor-switch intent, Claude generates concrete action items for marketing and sales leaders. Evaluates financial health and regulatory risk. This track monitors target pricing changes, extracts alternative data signals (like headcount velocity and news sentiment), and scrapes regulatory databases to generate a comprehensive 0-10 financial risk score. Acts as an autonomous threat hunter. It monitors threat surfaces (data breaches, CVEs, GitHub exposures), tracks compliance changes (CISA, NIST, GDPR), and conducts vendor risk assessments. Crucially, it features an AI agentic loop that autonomously investigates high-severity threats and triggers automated webhook alerts for critical vulnerabilities. Once the three parallel pipelines complete their analysis, a specialized synthesis agent correlates the findings. It produces a high-level executive summary, identifies the top three cross-departmental priorities, and maps hidden connections between the tracks (e.g., identifying that a sudden hiring surge is linked to a vulnerability in an exposed tech stack). Users can interact with the platform via a cyberpunk-themed Web UI, a robust REST API (allowing modular execution of individual tracks or full reports), and webhook integrations for real-time critical alerts. Bright Data Infrastructure: Relies heavily on the SERP API, Web Unlocker, Scraping Browser, and Web Scraper API to bypass anti-bot measures.

OmniMind: The Next-Generation Full-Stack Data Intelligence Mind OmniMind is a high-performance "data mind" application meticulously engineered by carolinesedurica-ux. Designed to serve as a hybrid internal AI assistant and sophisticated data visualization dashboard, this platform transforms the foundational google-gemini/aistudio-repository-template into a production-ready enterprise solution. Built with a strict TypeScript architecture for maximum type safety and reliability, OmniMind is currently live and optimized at omni-mind-delta.vercel.app. OmniMind moves far beyond simple prompting by deeply integrating Gemini models with a professional-grade Firebase infrastructure. It utilizes a secure backend featuring custom Firestore rules, robust user authentication, and structured application blueprints to ensure that data integrity and security remain paramount in an AI-driven environment.

Six Matches. One leaderboard. Build on AMD, earn points and compete for prizes.

🌎 Hybrid Hackathon | 💻 Online Build: 12-17 October 2026 | On-site phase: 17-18 October 2026 To participate in the hackathon, simply click Sign up with AMD. If you are not already a member of the AMD AI Developer Program (ADP), you will need to create an account. The AMD Developer Hackathon: ACT III is a hands-on event for developers, founders, engineers and builders who want to push what’s possible with AI on real infrastructure. 📍 If approved for the on-site experience, you'll also have the opportunity to showcase your project live. Travel and accommodation will not be covered.

🌎 Online Hackathon | 💻 Dates: 28 August–4 September 2026 | 🏆 $6,000 Prize Pool Build AI trading agents on Alpaca — autonomous agents and trading apps using Alpaca's Trading API, MCP server and CLI. 🌍 Fully online hackathon. Join and build from anywhere in the world.

📅 June 12–19, 2026 • June 12: Kick-off Stream • June 12–19: Online Build Phase • June 19: Project Submissions Ends 🌍 Fully online hackathon. Join and build from anywhere in the world. 🤝 Build practical enterprise workflows where AI agents communicate, coordinate, exchange context, and complete work together. 🧠 Powered by Band and Codeband, built for collaborative AI systems and agent orchestration. 🏆 Prize Pool: $10,000+ 🧑‍💻 Apply now to build the next generation of enterprise multi-agent systems.