Legacy code analysis
AI maps an unfamiliar codebase fast. I verify what actually matters.
Modernized, not rebuilt
Same system, stronger foundation
Evolve the system you have. No risky rewrites.
Where AI speeds things up — without cutting corners on quality.
AI maps an unfamiliar codebase fast. I verify what actually matters.
AI drafts data-migration scripts. Reviewed line by line before they touch production.
AI helps generate the first wave of tests on untested code. I decide what's worth testing.
AI drafts a module-by-module plan. I set the sequence and the guardrails.
AI proposes the upgrade diffs. I own every merge.
Most systems don't need a rewrite. They need a clear path forward, taken one safe step at a time.
Most production systems don't need a rewrite. They need a clear, incremental path off legacy tech — without stopping the business while it happens.
Code archaeology
Faster orientation in an unfamiliar codebase, so the plan is based on what's actually there.
Migration scripts
Faster first drafts for data migration. Every script reviewed before it runs against real data.
Test coverage
A faster path from zero tests to real coverage, so changes are safe to make.
A university faculty's core systems — legacy Perl and Oracle — modernized to Laravel, PostgreSQL, and Kubernetes. No downtime. The team kept shipping the whole time.
A B2B platform's front end rebuilt from CodeIgniter to Laravel, specifically to make it testable — coverage went from 0% to 50%, 2,400+ tests. Two years later it was picked for global rollout.
The tools I use daily.