I want to help build the
database that just works.
I'm Scott Shidlovsky — 12+ years shipping production web apps, the last seven scaling Cyberstrong from post-MVP to ~$10M ARR on React, TypeScript, Postgres, and AWS. The dashboards, APIs, and admin tooling are the work I want to be doing every day.
AI helped me draft this page. The claims, structure, and voice are mine — every line was reviewed and chosen deliberately.
// about
The short version.
I'm a hands-on VP of Engineering at Cybersaint Security. I architected the Cyberstrong GRC platform from post-MVP to ~$10M ARR leading a team of five — and I still write code every day.
My favorite work sits exactly where Surfaces does: the dashboards customers spend their day in, the REST APIs they automate against, the admin tooling operators rely on. The database is the product — Surfaces is what makes a serious database approachable, observable, and safe to operate. I've shipped that kind of supporting surface area at Cybersaint end-to-end — UI, API, ingestion, infra — for seven years.
// why-ic-why-now
Why IC, why now.
After seven years building and running a platform org, the work I still look forward to most is the IC work — the tricky React refactor, the Postgres plan that should be 10× faster, the API shape that quietly changes everything for a customer.
Management is part of my current job; it isn't the part I'm optimizing for. I want my next chapter to be the platform itself as the product, with my hands directly on it.
I'm not running from leadership — happy to mentor, review, and pull weight in design conversations. I just want to write the code and ship the surface.
// how-i-work
Ownership. Curiosity.
Ownership.
The thing I want from an engineering role is the whole loop — design, build, ship, incident, customer feedback, tweak, ship again. A feature isn't done when the PR merges; it's done when the people using it can feel that it works.
The "P99 engineer" bar reads to me as exactly this. I thrive when I own a surface end-to-end and stay on the hook for it — not just through the launch, but through the follow-on tweaks and the 2 AM page.
Curiosity.
I read source. When something behaves unexpectedly I'd rather trace through the Postgres planner, step into the framework, or skim a library's GitHub history than guess. LLMs are a great drafting partner; they're a bad excuse to skip the understanding.
The payoff compounds. Years of doing that homework leave you with the right instinct when a query plan goes sideways, a decent guess about the memory behavior of a Python service you've never run, or a sense of why a library made the API choice it did. Those small accumulated nuggets are usually what move an incident from "I'm stuck" to "I think I see it."
// shipped
Things I shipped.
Short receipts. Bigger system stories on request.
- Postgres ingestion redesignApplication-layer diffing + hashing strategy, with a large cut to write volume.
- 30M-point asset scoring pipelineConfigurable cadence shipped as a paid add-on driving expansion ARR.
- On-prem Kubernetes deploy pathCloud-native platform adapted to customer-managed clusters under a contract-critical deadline. Secured a $3M deal.
- Telemetry → risk signalsEmbeddings + LLMs to group large volumes of external telemetry (DLP alerts, attack paths) into actionable risk findings.
- Litesport billing lifecycleGo + Braintree, full lifecycle (discounts, pause/resume) on a subscription product at ~$100k/month MRR.
Let's talk.
I'd love to dig into where app.planetscale.com is heading, what operators struggle with most in the admin tooling, and the surface area you wish you had more hands on. Happy to jump on a call or meet for lunch.