Work
CHP Report Automation
End-to-end system that retrieves traffic collision reports for PI law firms — from email intake to AI phone calls to automated portal submission.
0 firms · 0 users · 0+ requests
Next.js · Convex · Playwright · VAPI · Twilio
→ Read case study
Knowledge Extraction Engine
Turns expert knowledge into structured, machine-readable rules through AI-conducted interviews. Hundreds of operational rules extracted and versioned.
0 rules · 0% coverage · 0 categories
Next.js · Convex · Claude API
→ Read case study
Multi-Agent Case Analysis
Parallel dispatch system where specialized AI agents analyze different aspects of a case simultaneously. Results aggregated and severity-ranked.
0x throughput · HIPAA-compliant · hybrid cloud+edge
Next.js · Convex · Claude API · Filevine API
Always-On Agent System
Multi-agent system on dedicated hardware that runs 24/7. Handles messaging across channels, spawns sub-agents, routes to the right model per task.
Runs daily · multi-channel · persistent memory
Mac Mini M4 · Multi-Provider AI
→ Read case study
Stack
AI / LLM
- Claude API
- OpenAI API
- Multi-Agent Systems
- RAG & Embeddings
- Prompt Engineering
- Agent Orchestration
Stack
- TypeScript
- Python
- Next.js
- React
- Node.js
- Convex
- Tailwind
Automation
- Playwright
- Voice AI (VAPI)
- Twilio
- Browser Automation
- Scheduled Jobs
Infra
- Vercel
- Fly.io
- Docker
- Git
- Dedicated Hardware
Compliance
- HIPAA
- CA SB 1518
- Human-in-the-Loop
- Audit Logging
- Encryption at Rest
Industries
Personal Injury Law
Thousands of decisions per week, buried in paperwork. I build systems that track cases, hit deadlines, and flag what matters — automatically.
Operations & Field Sales
Reps in the field, leads going cold, managers drowning in data. I build the intelligence layer that keeps your pipeline moving.
Admissions Consulting
Hundreds of students, dozens of deadlines, zero margin for error. I build systems that manage the chaos so consultants can focus on the students.
I started Manager Matt after spending 3 years going all-in on AI — not as a side project, but as my entire focus. I was in the first 1% of ChatGPT users. When the APIs opened up, I started building. When agents became a thing, I was already shipping multi-agent systems.
Most of my work is in legal tech. I understand the domain because I spent years in the personal injury industry before I started writing code. That combination — knowing how the business actually works AND being able to build the automation — is what makes my systems different from generic AI tooling.
I track every model release, SDK update, and framework change the day it drops. This is what I do every single day.
— Matt Sanchez, Founder