Pavlo Savushkin

Senior Full Stack Engineer - Laravel, Vue.js

Pavlo Savushkin, Senior Full Stack Engineer (Laravel, Vue.js)

Available for a new contract

Results you can put a number on.

I build systems from the first schema to production servers and own the technical decisions a business runs on. Four of them from zero to production in 10+ years. I build systems from the first schema to production servers. Four of them from zero to production in 10+ years.

Remote B2B contractor · Novi Sad, Serbia (CET/CEST) · US East or US West overlap

Recent client results

See more
  • 7x faster page load
    2.5 min to 20 s
  • ~$18k saved a year
    on infrastructure
  • ~1M transactions a month
    payments portal

pavel.savushkin.mix@gmail.com · I reply within one business day

When to call me

Results

Recent results from client systems.

  1. 01

    7x faster page load

    A sports analytics platform (via Svitla Systems, Inc.) opened its main page in 2.5 minutes on a cold start. I rewrote the legacy PHP code and the Elasticsearch layer; the page loaded in 20 seconds. Going further meant reworking the microservice architecture, which was outside the engagement.

    via Svitla Systems, Inc.

  2. 02

    ~$18k/year saved

    Same platform. I rewrote the MySQL and Elasticsearch configuration and optimized the code that works with them, so both moved to smaller instances, saving about $18k a year.

    via Svitla Systems, Inc.

  3. 03

    CRM built from zero

    A CRM for a certification company (NDA), built from scratch. I designed the data model and the architecture around the business needs, built the infrastructure and serve as tech lead. Live in production since launch.

    NDA

  4. 04

    1.5 h to 8 min per release

    For a US client (Upwork) I rebuilt CI/CD on GitHub Actions, Ansible and Docker Swarm. Releases went from 1.5 hours to 8 minutes, with zero-downtime deploys.

    Upwork

  5. 05

    3 weeks to 2 days per site launch

    An iGaming backend (NDA), built test-first. I moved the application core into custom pluggable packages that every new site reuses, so a launch took 2 days instead of 3 weeks (4 sites launched).

    NDA

  6. 06

    5x traffic peaks, zero downtime

    Same backend. I set up automatic scaling for the busiest campaign launches: the system grew with the traffic and shrank afterwards, absorbing 5x peaks without a minute of downtime or paying for idle servers.

    NDA

More from 10+ years

  1. Payments reporting portal for 100+ merchants and 5 payment processors, ~1M transactions a month, built from the ground up as the sole developer.
  2. Real-time vehicle-tracking system with WebSockets and role-based access, built from zero.
  3. Fault-tolerant server cluster across several hosting providers for an iGaming backend, designed from zero.
  4. Test-driven development introduced on two teams.
  5. Docker images and CI/CD pipelines on five engagements.

How I work

Core stack

10+ years of PHP and Laravel at the core.

Backend
PHP, Laravel (Lumen, Horizon, queues), REST APIs, Go for side projects
Data
MySQL, MariaDB, PostgreSQL, Redshift, Redis, Elasticsearch
Frontend
Vue.js, Nuxt, Quasar, Vuetify, TypeScript
Infrastructure
Docker, Docker Compose and Swarm, Kubernetes (AWS EKS), Terraform, Ansible, AWS (EC2 auto scaling, SQS)
Delivery and observability
GitHub Actions, GitLab CI, Grafana Loki, Prometheus, Bugsnag, Sentry

Terms

Contract

B2B contract with my registered business in Serbia, invoiced monthly. 3 months or longer, full-time or close to it. Start within a week of signing.

Hours and time zone

Novi Sad, Serbia, CET/CEST. 4 hours of US East overlap by default; I shift my hours to match US East or US West when the client needs it.

Code and accounts

Your code and accounts stay yours from day one.

Pilot

A paid 1-2 week pilot on request, so you can judge the work before a longer contract. No unpaid test tasks.

Quality

Engineering defaults

Architecture agreed with you before the first line of code.

Tests for every feature and fix; identical environments from laptop to production.

A second pair of eyes on every change. I propose the tooling; the final call stays with you.

Delivery

Every change goes through the pipeline and a staging copy of production. Releases are zero-downtime with a one-command rollback. A new developer is productive after one command.

Operations

Monitoring and alerts from the first release, so I see a problem before you do. Infrastructure and backups as code, restores tested. Secrets and access kept to the minimum.

Process

English

Fluent English. Every project so far has run in English with clients in the US, Canada and Europe: daily calls, code reviews, written decisions.

Calls and demos

Calls and demos on request, at a time that suits you.

Spec-driven development

Every task starts from a written specification and a plan agreed with you, so we both know what done means before the work starts. Where you are fine with it, AI coding agents work inside my own pipeline: they follow the specification, write tests first and pass automated quality gates, and nothing reaches your product before my own manual review.

Tooling

That pipeline is built on my engineering practice and on the best practices built into the tools it runs on: custom skills, hooks and automated checks keep every change inside those rules, so the result is under full control before it reaches my own manual review and your product.

Experience

7 roles since Oct 2015, from backend engineer to tech lead, newest first

  1. Lead Full Stack Engineer

    - present

    Proxify (contract)

    Greenfield CRM for a certification company: architecture, data model and API, deployments, CI/CD, backups and monitoring; tech lead of the engineering team.

  2. Senior Backend Architect

    -

    company under NDA, iGaming (contract)

    iGaming backend built test-first: new-site launch from 3 weeks to 2 days, automatic scaling that absorbed 5x traffic peaks with zero downtime.

  3. Senior Full Stack Engineer

    -

    Svitla Systems, Inc.

    Sports analytics platform: main page 7x faster (2.5 min to 20 s), ~$18k a year saved on database and search infrastructure, TDD introduced on legacy code.

  4. Senior Backend Engineer + Consultant

    -

    Hyuna International· part-time, in parallel with Svitla

    Rebuilt the merchant and payment processing core as the sole architect and developer; 5 payment processors behind one interface; payments made reliable with retries, refunds and idempotent requests.

  5. Senior Backend Engineer

    - present

    Upwork· freelance, short engagements between contracts

    CI/CD and zero-downtime releases for a US client: release time from 1.5 h to 8 min, one-command developer setup, identical environments from laptop to production. Also a real-time vehicle-tracking system with WebSockets and role-based access, built from zero.

  6. Software Engineer

    -

    Hyuna International

    Payments reporting portal (v1) for a Canadian client, built from scratch: onboarding, financial reports, payouts, reconciliations, exports and dashboards for 100+ merchants; frontend moved to a decoupled Nuxt app.

  7. Backend Engineer

    -

    Hyuna International (formerly JoMedia)

    Same company 2015-2023: payments reporting portal for 100+ merchants and 5 payment processors, ~1M transactions a month, built from the ground up; v1 and v2 as the sole developer.

    Refactored the billing core of a 20k+ line monolith serving 100+ merchants: database queries per billing operation from hundreds down to single digits.

Download CV (PDF, 66 KB)

PDF updated Nov 2026 The PDF is a one-page summary; LinkedIn has the full history.

Contact

Email is the best way to reach me. LinkedIn and Telegram work too, or send a message from here. I reply within one business day.

Paid pilot first, your code stays yours, no unpaid test tasks.

Loading the form...