Peter Gerhat

I came into mobile from the product and research side — which is an unusual path for someone who now spends most of their time in data pipelines and instrumentation code. That background means I think about analytics differently than engineers who started in infrastructure. The question I always start from is not "what can we measure?" but "what decision needs to be made, and what's the minimum data required to make it well?"

The pattern I've seen across a decade of work in PropTech, FinTech, public sector, and aviation is always the same: mobile teams build excellent products, then bolt on analytics as an afterthought. Events get named inconsistently, pipelines run without SLOs, dashboards multiply without ownership, and product managers stop trusting the numbers. The data exists — it's just not usable.

My work is to fix that at the architecture level, not the dashboard level. That means instrumentation-first development, schema governance, pipeline reliability engineering, and building the habits and tooling that keep data trustworthy as the product evolves.

Beyond analytics, I build offline-first React Native apps for field teams where connectivity can't be assumed, integrate LLM features with the instrumentation needed to measure them honestly, and design IoT telemetry pipelines for connected products. The common thread is end-to-end ownership — from the first SDK call to the decision that gets made in the weekly product review.

Skills & Stack