Three years out of law school, I picked up a statewide wiretap prosecution. A week later, I was trying a double murder.

The State had teams. Investigators, analysts, a war room, and time. I had a laptop, a calendar I was afraid to open, and tens of thousands of pages of discovery — terabytes of recordings, reports, and transcripts — and no one to hand any of it to.

There is a particular fear that arrives at 2 a.m. with a case file open and a trial date closing in: the certainty that the one thread that unravels the State's whole case is somewhere in those pages — and the terror that you won't find it in time.

What I learned in those months is that a criminal case isn't only an argument. It's a data problem. Every officer has a history. Every witness has a prior statement. Every precedent has a counter-precedent. Every timeline has a crack. One person can't hold all of it at once — and the case turns on exactly the details a tired human drops.

So I started building. First a script to surface the contradictions I knew were buried in there. Then a way to map each charge to the evidence that did — and didn't — support it. Then a way to see, at a glance, where the case was weak and what to do next. A script became a system. The system became something I talked to like a second chair.