AXIOM

Inspiration

Companies remember what they decided, but often forget why. We wanted to preserve the reasoning and context behind important decisions before they disappear into meetings, Slack, emails, and documents.

What it does

AXIOM captures decisions, reasoning, evidence, alternatives, and assumptions into a Decision Memory Graph. Users can ask questions like “Why did we choose PostgreSQL?” and get the original context. Its Decision Re-evaluation Agent also identifies when past assumptions no longer match reality.

How we built it

We built AXIOM as an AI-powered system combining LLMs, document and conversation analysis, structured decision extraction, a knowledge graph, and semantic search. Decisions are connected to people, evidence, alternatives, assumptions, and outcomes for contextual retrieval and re-evaluation.

Challenges we ran into

The biggest challenge was turning unstructured organizational conversations into reliable, structured decisions while preserving their context and relationships. We also had to design a system that could distinguish actual decisions from ordinary discussions.

Accomplishments that we're proud of

We built a working concept that goes beyond a traditional chatbot. AXIOM can capture organizational memory, explain the reasoning behind decisions, and identify decisions that may need to be reconsidered.

What we learned

We learned that organizational knowledge isn't just information - context matters. A useful AI system needs to understand not only what happened, but the reasoning, assumptions, and evidence behind it.

What's next for AXIOM

We want to make AXIOM continuously learn from an organization's data, integrate with tools like Slack, email, and project management platforms, improve decision-risk detection, and provide proactive alerts when important decisions become outdated.

Built With

• next.js
• python