AI Tinkerers Hackathon Project Gallery - Humans-in-the-Loop Agents Hackathon
AI Tinkerers - San Francisco
Final round winners have been announced. View Results

Entries

Browse submitted projects, teams, and winners.

26 Entries
Entry Description Team Status
CallFlow: AI-powered confidence-based routing with real-time verification and monitoring, auto-resolving routine inquiries and escalating complex cases for human approval. 1st Place
Seeking hackers to build an AI-driven, edge LLM companion toy using MQTT/IoT/ETL/TypeScript to comfort children with leukemia. 2nd Place
Can LLM agents guide laboratory experiments and correct mistake 3rd Place
Post-surgery care companion Finalist
Voice assistant that sends emails, Slack DMs, creates ClickUp tickets, and schedules Google Calendar events — voice or text control. Finalist
Big HONC deploys AI agents at {NameOfEvent} to analyze codebases and generate reusable end-to-end API testing collections. Finalist
Product roadmap based on GitHub Finalist
Build an AI-driven property-management platform that automates tracking and KPIs—Product Manager/ex-agent/investor seeks full-stack engineer to join. Submitted
BAN CARS. Using state-of-the-art VLMs and human-in-the-loop realtime feedback, help people reimagine their cities free of cars. Cityscape segmentation + inpainting ➡ ✨magic✨ Submitted
Voter Guide Submitted
Traditional dev docs help humans/SEO but are invisible to LLM-driven coding because LLMs can't follow logical chains. Submitted
Use Gemini to identify which company documents address a regulation and deliver a clear compliance judgment. Submitted
AI meeting assistant that provides real-time screen understanding and contextual memory for technical meetings with shared code, diagrams, and docs. Submitted
Daisy connects seniors to local, tech-enabled interest communities, closing access gaps and reducing loneliness. Submitted
Date-X automates personalized dating and business meetings—suggesting plans, booking reservations, tracking budgets—powered by Google Vertex AI and Gemini. Submitted
Team consisting of Joe Hewett (CTO, Entropy Labs; BSc Warwick) and David Mlčoch (Co‑founder/CTO, Trident AI; MSc Edinburgh), building LLM-powered fraud detection and agent supervision. Submitted
Using ai to write Submitted
Team led by an Engineering Lead and Talentspace co‑founder (Handshake), HPI alumnus — React, Python, generative UI, agents, Cloudflare Durable Objects. Submitted
How do you make it easy for AI to select the right app API and show a UI to the user for confirmation? Submitted
Team consisting of a Meta data engineer (B.Eng CSU), a CBG ML engineer (MSc Ben‑Gurion), Astra founder and Natera QA — Python, MLOps, NLP, Docker/K8s. Submitted
Building small red‑teaming agents to probe a larger model, producing playful commentary on each agent's attack actions. Submitted
Building distribution tools for startups to grow on TikTok - mass generate AI slideshows + videos, experiment with different creatives Submitted
AI résumé summarizer highlights role-fit skills, lets recruiters filter and prioritize criteria to streamline evaluations while preserving human judgment. Submitted
Team led by a YC F24 founder/CEO (HumanLayer) with 11+ years building Python/Go/JS AI agent tooling, Kubernetes, and HITL workflows; UChicago Physics. Submitted
team biblahbloob is working on a prompt poisoning detector. Submitted
Agents that monitor Telegram channels for disinformation, flagging potentially false or misleading posts, images, and videos for further review Submitted