Shipps

Inspiration

We were inspired by the growing tension in global logistics: companies want faster and cheaper routes, but governments demand compliance with strict carbon regulations. Most tools either say “non-compliant” or force costly detours. We thought—why not let AI negotiate sustainability in real-time?

What it does

Simulates a debate between AI agents (Route, Carbon, Policy, Optimizer) to decide the best logistics path.
Turns compliance from a blocker into a solution by auto-purchasing carbon credits when limits are exceeded.
Produces a transparent conversation log, so users see how every decision was reasoned—not just the final result.
Balances cost, speed, and sustainability dynamically, instead of forcing one trade-off.
Reframes carbon credits as a financial tool, making sustainability part of the optimization—not an afterthought.

How we built it

We designed a multi-agent system where each agent has a specialized role:

Route Agent → Finds efficient and cost-effective paths
Carbon Agent → Calculates emissions for every route
Policy Agent → Validates compliance against regional rules
Optimizer Agent → Mediates between trade-offs to finalize the best decision

To showcase this, we built a pipeline using LangChain/LangGraph and mocked a Carbon Marketplace API. Judges can actually watch agents “debate” and settle on an optimal route.

Challenges we ran into

Getting multiple agents to communicate without looping endlessly or reaching deadlock
Balancing realism (market APIs, compliance rules) with hackathon-time constraints
Designing an engaging demo that’s both technically impressive and easy to follow

Accomplishments that we're proud of

Built a working multi-agent system where AI agents debate and agree on routes.
Integrated a mock Carbon Marketplace to auto-purchase credits for compliance.
Designed an engaging demo that makes AI decision-making transparent and interactive.

What we learned

Multi-agent systems are powerful but tricky—getting agents to collaborate without looping required careful orchestration.
Compliance isn’t just binary; real-world sustainability requires flexible tools like carbon credits, offsets, and financial trade-offs.
A transparent agent debate log helps build trust in AI decision-making by showing why a recommendation was made.

What's next for Shipps

Our next step is to evolve Shipps into a full-fledged carbon-credit marketplace where AI agents don’t just recommend—they act. Agents will be able to execute carbon-credit trades directly, while employees provide human oversight and intervention to ensure accountability. The goal: a scalable logistics platform that blends automation with human judgment to make global shipping both profitable and sustainable.