agents that sign up for services, verify emails, and pay. x402 + browser use + agentmail.
built at browser use web agents hackathon at y combinator hq
multimodal temporal/spatial memory and verification for vlm outputs. gemini + sam + depth + colmap.
ironsite prize + best use of solana @ hacktech 2026
claude does diligence on a github repo, then settles tokenized equity on xrpl with crypto-enforced vesting.
1st overall + 1st sui foundation track @ b@bhacks 2026
prompt-driven video editor for localized edits with continuity across shots. clip + sam 2.1 + ffmpeg.
2nd overall @ citrus hacks 2026
personal claude code daemon that runs 24/7 on a vps with discord, telegram, voice, cron, and multi-session thread isolation. bun + ts.
agentic context + finance layer for nonprofits. one chat across 13 connectors (bloomerang, salesforce npsp, m365, quickbooks, zoom, etc.) with cited answers + immutable audit log, plus two onchain rails on solana — http 402 agent donations and cause coins via meteora bonding curves.
hackdavis 2026 — winner, best use of solana.
embodied desktop ai coworker - swipe-based persona discovery, nia semantic memory, sms + discord fallback, and a kawaii electron/r3f room running codex in-world.
built at nozomio hackathon @ ef sf (nia + tensorlake always-on agents track)
openai codex skill repo turning videos into fully trained object detection models. yolo + codex.
top 5 ($10k) @ openai codex hackathon, demoed to sam altman + greg brockman
ai gaming companion you can queue up with. yolov8 + depth estimation + livekit.
$250k winner at pump.fun build-in-public hackathon, pitched live to tim draper
voice-to-3d spatial learning platform with gaussian splat rendering. react + gemini + marble + elevenlabs.
president's pick + mlh best use of elevenlabs @ sb hacks xii
ai agents compete to build react components in real-time. react + letta + livekit + sui + three.js.
top 3 letta agent track at cal hacks 12.0
turns research papers into 3blue1brown-style explainer videos. next.js + fastapi + gemini + manim.
originated at nvidia ai agent hackathon, rebuilt for public demo
natural-language search for github devs. ranks 60% gemini judgment + 40% raw github metrics.
top 10 @ b.e.l.l.e sf ai hackathon
