Agents Assemble - The Healthcare AI Endgame
Devpost hackathon: Agents Assemble - The Healthcare AI Endgame. Imported automatically.
- Year
- 2026
- Winning projects
- 13
- Participants
- 4,305
- Prize pool
- $25,000
- Platform
- Devpost
Winners
1st Place
- LookCloser1st PrizeA multi-agent safety net for incidental imaging findings, because 1 in 5 follow-ups never happen, and no rule engine can read a visit note to know why.Python
2nd Place
- LoopGuard Passport — Governed Clinical Loop Closure2nd PrizePrevent the diagnostic harm that begins the moment a follow-up falls silent.Python
3rd Place
- Ctrl+Alt+Heal3rd PrizeAI-powered clinical gap detection that finds the diagnoses hiding in your charts, so no condition goes uncoded and no revenue goes uncaptured.DockerFastAPIGoogle Cloud
Honorable Mention
- Aegis Hindi Consent CompanionHonorable Mentions**India's AI-generated perioperative informed consent in Hindi — patient-specific, clinically grounded, and built for the real communication gap in Indian surgiDockerPython
- Agentic Multidisciplinary Tumor Board (AMDT)Honorable MentionsOrchestration that uses A2A agents to synthesize unstructured FHIR pathology & radiology data into instant Tumor Board briefs. We eliminate hours of manual prepPython
- Discharge CoordinatorHonorable MentionsAutomates discharge coordination — reads clinical notes, validates CMS rules, and generates Standard Written Orders in seconds instead of hours.Python
- FHIR ForgeHonorable MentionsTransforms clinical notes, discharge summaries, and referrals into validated FHIR R4 resources with RxNorm, ICD-10, and LOINC grounding, clinician review, interTypeScript
- MaternalGuard: MCP for Maternal Risk on SHARP+FHIRHonorable Mentions5-tool MCP server plus A2A orchestrator that gives healthcare AI the full mother-baby picture from any FHIR R4 workspace: maternal risk, lab trends, SDOH, ACOG-TypeScript
- NEST: Newborn & Maternal Safe TransitionHonorable MentionsFragmented charts cost lives. NEST saves mothers and newborns by orchestrating 5 AI agents that synthesize complex FHIR data into a unified, evidence-backed disGeminiPython
- PediRounds — AI Co-Resident for Pediatric Morning RoundsHonorable MentionsThe safety net between overnight and attending rounds. 4 AI agents catch deterioration, flag unsafe doses, generate I-PASS handoffs, and produce bilingual parenFastAPIGeminiPython
- Pre-Round BriefHonorable MentionsAn AI chief-of-staff for hospital physicians, a proactive PromptOpinion agent that delivers a prioritized morning briefing of every patient on the doctor's paneDockerOpenAI
- RootCauseHonorable MentionsA FHIR-native multi-agent AI system that detects why dental patients stop coming in and automatically generates personalized re-engagement plans, orchestrated a
- TrustedRisk Care EngineHonorable MentionsA2A federation that decomposes one clinical prompt into a multi-specialist consultation: 16 sub-agents, 257 workflows, narrated dispatch, calibrated abstention AnthropicDockerFastAPI