Skip to content
EK

Search

About

Egor Kravchenko

Egor Kravchenko

Senior Full-Stack & Mobile Engineer

Antalya, Turkey · Remote only · UTC+3

Open to senior / staff roles · full-time or contract

I'm a senior full-stack and mobile engineer. I own production systems end-to-end — the API, the data model, and the React Native or web front end that ships them — rather than a single slice of the stack. Right now I'm the founding engineer at Medical Avenue, a medical-tourism platform, where I took over an inherited codebase and have been the sole engineer since; before that, almost five years at NayaOne, an FCA-regulated fintech platform, where I grew from senior engineer to team lead.

Experience

Founding Engineer (Contract)

Medical Avenue · Jan 2026 – Present

Took over a medical-tourism platform from a departed agency team and have been the sole engineer since — owning the React Native patient app, the backend, and the design system end-to-end.

  • Cut the inherited React Native app from a 1.2 GB memory peak with OOM kills on real devices to a stable ~350 MB by flattening its navigation architecture — verified on a 2018 Android with 3 GB RAM.
  • Added ~70 REST endpoints (Node.js / Prisma / PostgreSQL) for the v2 mobile API on top of an inherited ~120-endpoint platform now under my sole ownership.
  • Overhauled auth: slim JWT payloads, a refresh flow that verifies expired-token signatures with machine-readable error codes, Redis-backed sliding-window sessions, Google/Apple OAuth, RBAC, and rate limiting.
  • Built a secure file backend on S3-compatible object storage (Cloudflare R2): presigned upload → confirm → download → delete, a default-deny access registry, and append-only audit logging enforced by PostgreSQL triggers.

Senior Full-Stack Engineer & Team Lead

NayaOne · May 2021 – Dec 2025 · London

Enterprise SaaS platform serving 80+ financial institutions in FCA-regulated cloud environments for AI and fintech experimentation. Grew from senior engineer to team lead of five.

  • Delivered the launch module behind the FCA–NVIDIA AI Sandbox partnership end-to-end under a one-week deadline.
  • Architected automated sandbox provisioning — queue-based Terraform orchestration across AWS, Azure, and GCP — cutting enterprise deployment from days to under one for 95% of projects.
  • Cut GraphQL hot-path latency 250ms → 75ms (−70%) for 300+ daily users by refactoring 80+ queries with raw SQL joins and batched cross-database resolvers.
  • Built a session-analysis pipeline over 35K+ product sessions that exposed a 70% vendor-onboarding drop-off, then led the 5-person redesign that cut it below 5%.

Full-Stack Engineer

Synergy University · Oct 2017 – May 2021 · Moscow

Built a full-stack loyalty platform (Node.js / Vue.js) with enterprise SSO (Keycloak) and mentored two junior engineers to mid-level within six months.

Outside work I run self-hosted AI infrastructure — local embeddings, hybrid retrieval, and agents over my own knowledge base — and write about the engineering decisions behind all of this: the tradeoffs I actually made, the naive solution that broke, and the lesson that generalizes.

Selected work from these roles is written up as case studies. Full history on LinkedIn.

Want to talk about a role or a system like these?