Piotr Karmelita@piotr_karmelita146

Staff preparing for a visit have to read several short documents and notice what is already recorded, what is missing, and what conflicts. VisitBrief keeps that judgment with the person. It is built for clinical staff, not for a patient looking for advice, and it prepares a draft from an allowlisted fictional chart. The staff member picks one of twenty built-in cases. There is no file upload and no free-form prompt, so the demo cannot accept a real patient record. One to three text PDFs are split into pages and source blocks. DeepSeek-V4-Flash on AMD Token Factory returns candidate facts. Radeon Cloud states that these catalog models run on AMD GPUs. The API keeps a fact only when its quote is inside the block it cites. If the model paraphrases a value, the screen shows the quote instead. Wording that is not in the assigned block is dropped and is not repeated. Duplicates are merged. A date that is not in the quote is cleared, so undated facts stay off the timeline. Two recorded doses of the same medicine are shown side by side and are not reconciled. A sentence that tells the model to ignore instructions or invent a diagnosis is dropped, and the case is flagged. Missing allergy documentation stays a review flag, not a finding of no allergies. The screen counts candidates that were kept, repaired, or dropped, and it labels the result as a draft for clinician review. There is no database. Request logs do not store document text. Held-out cases were not used to change the prompt. One pass on 15 October recalled 27 of 58 curated development facts (46.6 percent) and 34 of 43 held-out facts (79.1 percent).

UvicornForge AI is an AI-powered co-founder designed for hackathon teams. Enter a project idea along with real parameters such as team size, total funding, available time, target users, industry, and technologies (including AMD GPUs and Fireworks AI). The system generates a comprehensive, structured startup brief covering problem, solution, MVP scope, key features, demo scenario, business model, risks, go-to-market strategy, and more. It also produces ready-to-use artifacts including a pitch deck outline, full demo script, MVP checklist, and starter README. A key innovation is the Success Score (1-10) predicted by a custom MLP model. Unlike generic scores, ours is trained on a high-signal target engineered to respond to actual project parameters: larger teams, realistic funding, longer development time, and explicit AMD usage all meaningfully increase the predicted success. The frontend allows direct input of these values, and the feature mapper feeds them straight into the model along with an ambition factor extracted from the idea description. The backend combines Fireworks AI for high-quality text generation with a locally running PyTorch MLP (65 features) trained on an AMD-tailored dataset of 10,000 examples. We achieve strong validation performance (R² ≈ 0.85) while ensuring the score differentiates between weak and strong proposals. The modern frontend features live model metrics, smooth animations, and one-click downloads of all generated materials. Built specifically for the Unicorn Track, the project demonstrates practical use of AMD GPUs for ML inference and Fireworks AI for LLM generation in a real, useful product that helps teams create better submissions faster.

PM PLUS transforms project management from a passive, administrative burden into a self-optimizing, autonomous ecosystem. Traditional enterprise workflows suffer from constant status-chasing and reactive bottleneck detection, causing preventable delays and distracting leaders from high-level strategy. Our solution replaces static tools with an intelligent, multi-agent workforce. Powered by the Band coordination layer, these agents proactively manage project health, negotiate resource reallocation, and maintain deep historical context. By resolving systemic issues and optimizing workflows in the background, PM PLUS ensures that project managers transition from chasing updates to driving strategy. With human oversight maintained through a real-time, intuitive dashboard, PM PLUS delivers unprecedented efficiency, moving beyond mere task tracking to intelligent, automated project success.

Risk Horizon is a multi-step AI agent for supplier risk and market disruption intelligence. Companies often depend on critical suppliers, logistics providers, manufacturers, and raw material sources, but early warning signs of disruption are scattered across the public web. Risk Horizon helps teams detect those signals faster by using live web data to search for supplier-related news, advisories, regulatory updates, market commentary, and disruption reports. The agent collects relevant sources, analyzes them for risks such as delays, shortages, sanctions, lawsuits, price changes, factory issues, and geopolitical events, then generates a source-backed risk report. Each report includes a risk score, confidence level, warning signals, business impact, source evidence, and recommended actions. Risk Horizon is designed for procurement, supply chain, operations, and finance teams that need to act before supplier disruptions become expensive business problems.

LazyBob is an intelligent development accelerator created for the IBM Bob Hackathon, designed to help developers transform ideas into working solutions faster and more efficiently. The platform combines automation, AI-powered analysis, and modern development workflows to reduce repetitive work and improve productivity across the entire software development lifecycle. Powered by IBM Bob and watsonx AI technologies, LazyBob assists developers with code understanding, automated documentation generation, debugging, refactoring, and workflow automation. The system can analyze existing codebases, generate onboarding materials, create API documentation, and provide intelligent code suggestions using IBM Granite models. It also supports automated code reviews, quality scoring, security analysis, and AI-driven task execution. A key part of the project is the interactive web dashboard, which provides real-time analytics, workflow monitoring, code quality visualization, and AI usage statistics through a responsive and modern interface. By combining backend automation with advanced AI capabilities, LazyBob acts as a smart development companion that accelerates learning, coding, collaboration, and project delivery.

OpsTune is an AI-powered industrial incident analysis system designed to help operations teams understand and act on technical incidents faster. In industrial environments, operator reports are often written as unstructured text, making it difficult to quickly identify severity, root causes, evidence, and next steps. OpsTune transforms these raw reports into structured, actionable intelligence using a fine-tuned large language model running on AMD hardware. The system takes a messy incident report as input and returns a clear JSON analysis including severity level, incident category, likely root causes, supporting evidence, recommended actions, confidence score, and a concise operational summary. This enables faster triage, better decision-making, and more consistent incident handling. Our goal is to demonstrate how domain-adapted AI workflows can improve industrial maintenance, reduce downtime, and support frontline teams with reliable, explainable incident analysis.

The Problem: Moral Failure and Stagnation (Focus: Relatable) Open-source healthcare innovation is crippled by subjective recognition, slow development, and the moral failure of systems prioritizing profit over patient care. Developers lack objective metrics for contribution quality, leading to burnout and talent exclusion based on static credentials, not real-world impact (e.g., the 900x faster MCP Server). The Solution: QUBIC's Meritocratic Architecture (Focus: Mechanics) QUBIC is the solution. It is a hybrid, autonomous governance protocol built for the Nostromo Launchpad. Off-Chain AI Analysis: Our AI Judge (LLM-Powered) scores every contribution (code quality, documentation, clinical utility, security compliance) against a public, weighted rubric. On-Chain Integrity: A secure Oracle posts the composite score to the Qubic C++ Smart Contract DAO. Trustless Rewards: The Smart Contract registers the contributor with a Platinum, Gold, or Silver tier and automatically distributes UC tokens and weighted voting power. The Qubic Advantage (Focus: Technical Edge) QUBIC leverages Qubic's high-performance C++ architecture for fast, deterministic execution, essential for mission-critical healthcare governance. This ensures that the DAO remains a Meritocratic Democracy—where governance power is based on proven, life-saving contributions, not token wealth. QUBIC is the trust layer that funds and sustains the Ubuntu Patient Care Rebellion.

Introducing RETENX: An AI-powered multi-agent system designed to help enterprises uncover hidden risks across customer relationships, procurement processes, and financial operations before they escalate into costly problems. By continuously analyzing complex data streams, RETENX provides early detection of potential churn, operational bottlenecks, and financial vulnerabilities, enabling teams to act strategically and safeguard revenue. With advanced risk correlation across departments, RETENX not only identifies isolated issues but also highlights systemic patterns that could impact overall business performance. Its autonomous workflow execution automates routine interventions, freeing teams from manual monitoring while ensuring timely and effective action. RETENX delivers real-time alerts, actionable task assignments, and executive-level briefings, equipping decision-makers with the insights they need to protect customer relationships, optimize processes, and drive sustainable business growth. By turning reactive measures into proactive intelligence, RETENX empowers organizations to stay ahead of risks, reduce operational losses, and enhance long-term resilience in a dynamic market environment.

AuraCare Assistant is an intelligent, AI-powered platform designed to revolutionize the patient experience by tackling the administrative friction in healthcare. From tedious intake forms to the confusing process of finding the right specialist, our application automates routine tasks, provides intelligent recommendations, and empowers patients to take control of their healthcare journey. By streamlining the administrative workflow, AuraCare allows healthcare providers to focus on what truly matters: delivering exceptional care. Our vision is to create a healthcare ecosystem where technology enhances the human element, making every interaction smoother, faster, and more empathetic. We believe that by reducing administrative burdens, we can foster better patient-provider relationships and improve overall health outcomes.

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

🌎 Hybrid Hackathon | 💻 Online Build: 3–8 November 2026 | On-site Phase: 8 November 2026 📍 On-site Venue: Pierpont Place 163 W Pierpont Ave, Salt Lake City, UT 84101, USA ✨ Please note that travel and accommodation are not covered.

🌎 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 | 💻 Online Build: Sept 26 - Oct 5, 2026 A virtual, global hackathon. Build from anywhere in the world. Dark Factory is the official hackathon of WeAreDevelopers World Congress North America, presented by BAND, but you don't need to be in San Jose to take part. For one week, you build a software factory in BAND Desktop: a band of coding agents that plans work, implements it and checks its own results. 🎟️ Bonus for sign-ups: you are eligible for a free ticket to WeAreDevelopers World Congress North America in San Jose (Sep 23–25). If you do go, stop by the BAND booth: a surprise is waiting for you.

🌎 Online Hackathon | 💻 Month-long challenge | Dates: Sep 1–30, 2026 | 🏆 $10,000 Prize Pool ($5k cash + $5k in AAI credits) A month-long online challenge run by lablab.ai together with AssemblyAI, the Voice AI infrastructure company for builders. Every participant builds on AssemblyAI. You can join at any point — registration stays open for the whole build window, so you can start on day one or pick the project up halfway through and still submit. 🌍 Fully online hackathon. Join and build from anywhere in the world.