AI Studio: Demo Day
You must be registered and approved. Bring your QR code from Luma and your ID. Demos start at 9:50am. From noon the audience will have a chance to engage with the teams during the demo showcase. MIT or Harvard members should register with their MIT or Harvard email address.
AI Agents demo for the MIT AI Studio Course
We invite you to MIT AI Studio Demo Day where 40 agentic AI-first projects will present on stage in front of judges comprising VCs, AI experts, and serial entrepreneurs.
These teams have spent the semester building at the frontier of AI agent infrastructure and applications — tackling the foundational questions of how autonomous agents will transact, coordinate, trust each other, and operate in the digital and physical world. Projects span agent identity and reputation systems, programmable financial rails for machine-speed commerce, portable agent packaging, hardware discovery protocols, multi-agent orchestration layers, and marketplace mechanisms where agents autonomously discover, price, and verify work. On the application side, teams are deploying agents into healthcare compliance, biomedical research, education, food brokerage, CNC machining, trading, and human performance — while others are solving the hard trust and safety problems that emerge when agents act independently: cryptographic data isolation, governance of agent work cycles, transaction dispute resolution, and rigorous evaluation frameworks. Together, this cohort is building toward a future where AI agents operate as first-class economic actors — with the protocols, infrastructure, and guardrails to make that real.
Follow MIT policies for privacy and accessibility listed at https://aiforimpact.github.io. You are only permitted to attend the event on the 3rd floor from 9:30am to 1pm. We want you to leave immediately after the event as the building is not open to visitors.
Demo Day lineup:
AAX — Agentic ad exchange for matching and closing advertising opportunities in real time.
Agent Neo-Banking Rails — Programmable banking, settlement, and credit rails for autonomous AI agents transacting at machine speed.
Agentic Reputation Layer — A trust layer where AI agents earn reputation through real work.
Agents Army — The agents behind a CNC machining shop.
AI Agents for Food Brokers — A multi-agent system that automates food-broker workflows across sourcing, outreach, operations, and portfolio management.
arbiter — A transaction dispute and accountability layer for agent systems.
Articulate — An AI-powered oral exam platform that brings the productive struggle of articulation back into learning.
Beyond Answer Engine — An AI learning system that helps students understand concepts by forcing explanation instead of giving answers.
Bukti — A trust and verification layer for humans and agents.
CKC — A coordination layer for multi-agent systems that finds the right agents, ranks them, assigns work, and keeps humans in control.
Clawnection — An agent-assisted matchmaking platform where your agents can go on dates before you do.
Clawviyo — The relationship intelligence layer for AI agents.
DataX — A marketplace for small, structured datasets that help agents with retrieval-augmented tasks.
Debot — A shared knowledge base where AI agents search for answers, post questions when stuck, and verify what works.
Evaluations — A system for users and agents to build, refine, and store evaluations for LLM and agent systems.
Hivee — A collaborative workspace where specialized AI agents from different people and organizations work together on shared projects.
Jackfruit AI — Helping enterprises take control of their infrastructure and make it ready for AI.
Lattice — A system that turns a research hypothesis in plain English into a defensible engineering verdict grounded in physics, literature, and computation.
Lily — An AI-agent platform that helps people reduce unwanted behaviors by predicting high-risk moments and delivering real-time support.
MahaGuardian — Cryptographically enforced data-partition isolation for split-trust autonomous agent systems.
Merit — Credit infrastructure for the agent economy.
Moonshot — A multi-layer platform for buying and selling private-company shares.
Mycelium — Impact-weighted knowledge routing for enterprise AI agent teams.
Nexus for Federated Learning — An LLM-powered agent that autonomously manages federated learning with auditable decisions.
Nydus — A portable packaging layer for AI agents that lets users safely create, share, and migrate agent state without framework lock-in.
Octopus Protocol — A protocol for autonomous hardware discovery and control by AI agents.
Op-Era — AI operations research for mid- and large-scale companies in Latin America.
Personalized Wealth Manager — Infrastructure that lets retail users access private-wealth services and run customized trading strategies.
Pison AI — A text-message AI agent connected to wearables and apps that tracks, analyzes, and improves human performance.
statem8 — A blockchain-based registry where AI agents get verifiable identity, transparent financial history, and public reputation scores.
Terrain — Infrastructure for deploying AI agents to IoT devices by scanning networks, pairing devices, and managing deployment.
The Ghost Trader — An experiment in how far AI agents can run a profit-generating business, starting with digital-product trading.
Umoja — A persistent event-graph system where personal agents accumulate interaction memory across gatherings.
Vouch — A shopping network for agents based on trusted recommendations and purchase-history-aware suggestions.
WHITE ROOM — A governance layer for managing the work-and-rest cycle of agents.
ZeNi — The operational backbone for AI-agent teams.
About the course:
https://aiforimpact.github.io
This MIT course teaches students to build autonomous AI agents that plan, coordinate, and execute complex workflows across distributed web systems. Students learn agent architectures, inter-agent protocols, web automation frameworks, and methods for decentralized coordination, while mastering AI-native coding and multimodal integration.
Through hands-on projects, students build sophisticated agentic applications: from automated workflow orchestration to intelligent web scraping and dynamic API coordination. The course combines solid technical foundations with innovative product development opportunities in this emerging infrastructure layer. Students learn to design AI agents with global impact, validate ideas through AI-assisted research and interviews, and move working prototypes toward market-ready solutions.