Chronos AI: Legal and medical document chronology platform
By Colter Mahlum, Founder & CEO of Mahlum Innovations — published 2026-05-01.
Colter Mahlum architected and shipped this case study end-to-end. Mechanical engineer turned AI builder, based in Bigfork and Kalispell, Montana. 12 platforms built across healthcare, wellness, legal, wealth management, fitness, and consumer apps.
Turning complex record sets into structured, searchable patient timelines for law firms
App: Chronos AI | Industry: Legal Technology | Client: Legal document intelligence platform (designed for law firms handling medical record review) | Company size: Law firms handling personal injury, medical malpractice, and workers compensation matters | Duration: Initial build complete, ongoing development
Summary
A document chronology platform that ingests legal and medical record sets and produces structured event timelines, built on React, TypeScript, Node, and Supabase Postgres with pgvector for semantic search across document corpora.
The Challenge
Law firms handling personal injury and medical malpractice matters routinely receive hundreds to thousands of pages of medical records per case. Turning those records into a structured, searchable chronology is time-intensive attorney and paralegal work that can take days per matter and introduces inconsistency under deadline pressure.
Our Approach
- Built an ingestion pipeline that accepts multi-format document uploads and extracts structured events with dates, providers, findings, and diagnoses
- Used pgvector in Supabase Postgres for semantic similarity search across the full record corpus so queries find relevant records by meaning, not just keyword
- Designed a React and TypeScript frontend for reviewing, editing, and exporting structured chronologies
- Built a Node API layer with document parsing and event extraction pipeline
- Implemented per-matter organization with role-based access control for attorney and paralegal workflows
Outcomes & ROI
- Structured — Event extraction: Unstructured medical records transformed into dated, typed, searchable event records
- Semantic — Record search: pgvector enables similarity search across the full record set rather than keyword-only lookup
- Exportable — Chronologies: Structured timelines exportable for use in demand letters, depositions, and trial preparation
Reduces the time from record receipt to a reviewable chronology and makes the underlying record corpus searchable by meaning rather than exact keyword match.
Technologies Used
React, TypeScript, Node.js, Supabase Postgres, pgvector, Drizzle ORM
Key Takeaways
- pgvector makes a document corpus searchable by meaning rather than keyword, which changes how legal teams find relevant records in large files
- Structured event extraction is the foundation for any downstream legal analytics. Getting the schema right early saves significant rework later
- Role-based access control distinguishing attorney and paralegal permissions is necessary from day one in legal tooling, not an afterthought
Frequently Asked Questions
What document formats does the platform accept?
The ingestion pipeline handles multi-format uploads including PDFs and scanned documents. Events are extracted with structured metadata including date, provider, facility, and clinical category.
How does semantic search work across medical records?
Records are embedded using pgvector in Supabase Postgres. Search queries are also embedded and matched by vector similarity, so a search surfaces relevant records even when they use different clinical terminology than the query.
Is the platform in use at law firms today?
The platform has been built and is in active development. Client onboarding is ongoing.
Related Services
About the Builder
Mechanical engineer turned AI builder. Colter personally architected and shipped Chronos AI end-to-end — strategy, model development, and the production MLOps work — alongside 10+ other AI systems across manufacturing, wealth management, healthcare, and consumer apps. Read full bio · LinkedIn.
More Case Studies
- Axiom AI — A multi-agent AI orchestration platform that lets users deploy, coordinate, and monitor swarms of specialized Claude/GPT/Gemini agents from a single command center to autonomously execute complex software, research, and operational workflows.
- Axiom Trading Agent — An autonomous multi-vertical trading agent that finds, validates, and executes positive-EV trades on Kalshi prediction markets and Robinhood crypto end-to-end, with no human in the loop.
- Montana Athletic Club App — A member mobile application for a regional Montana health club, featuring Apple HealthKit integration for activity and health data sync, member account management, and class scheduling.
- Elysian Operations Platform (Time Warp + Nexus Wellness) — All-in-one workforce, payroll, wellness, and tax-compliance platform that replaces 8+ separate SaaS tools with a single unified system.
- Elysian Institute Website — A modern, SEO-optimized marketing website for a recovery, wellness, and longevity center, featuring 20 detailed service pages, transparent membership pricing, and direct integration with the client's external booking system.
- March Madness Commissioner — A full-stack commissioner dashboard that automates ESPN bracket league management — syncing picks, tracking payments, and generating AI-powered win projections — with a companion iOS mobile app.