Solutions & Integration Engineer

Systems that
work in reality.

I connect business software, data, and external services so everyday work becomes faster, safer, and less manual, from the first problem description to deployment and support.

Python / PHPBackend systems
Linux / DockerProduction operations
APIs / SQLIntegration & data
9 yearsE-commerce ownership

01 / Expertise

One engineer across the whole system.

You explain what needs to work. I find the systems involved, build the connection, put it into production, and remain responsible when something fails.

01

Connecting systems

Before: people copy the same data between applications or work around systems that do not communicate.After: the applications exchange the right data automatically through APIs, databases, or files.

  • APIs & external services
  • Spreadsheet & database imports
  • Validation, logs & safe retries
02

Keeping systems running

Before: when something fails, it is unclear whether the cause is code, a container, a database, or the network.After: the system is monitored, backed up, and easier to diagnose and restore.

  • Linux, Docker & Proxmox
  • Monitoring & troubleshooting
  • Deployments, backups & security
03

Improving legacy software

Before: manual steps and old code make every change slow and risky.After: repeatable automation and tests make work faster and changes safer to release.

  • Python process automation
  • Legacy PHP modernization
  • AI suggestions with human approval

02 / Selected work

Before, after, and measurable results.

These examples explain what was difficult, what I changed, and what now works. Private client code and business data stay private; only verified evidence is shown.

CASE / 01

Legacy system · Anonymized

Making a critical legacy system safer to change

Before: important business rules were spread across old PHP code, database rules, and scheduled jobs. Even a small change could break another process.After: I mapped the risky parts, removed obsolete files, and added automated tests, repeatable database changes, Docker services, and two controlled deployment environments. Changes can now be checked before release instead of relying only on manual testing.

PHPFastAPIPostgreSQLDockerPHPUnitCI/CD

Measured evidence

396 automated tests · 1,108 checks
29 documented releases
36.3% fewer files in the codebase
Verified in July 2026: zero failures and one credential-dependent test skipped. Business details remain private.
CASE / 02

Client data import · Anonymized

Different client spreadsheets, one importer

Before: every company sent CSV, XLS, or XLSX files with different columns and layouts. A new layout could stop the import and require a programmer to change the code.After: I built one import process that maps each file into a common structure, lets an operator correct the mapping, and remembers successful layouts for later imports. A bad row is logged instead of stopping the whole file, and the job can be stopped and resumed. If the company identity is uncertain, AI can suggest the most likely match, but a person must approve it before sensitive data is saved.

CSV / XLS / XLSXFlexible column mappingCompany registryHuman approvalError log

What now works

Client file → mapping → common format
New layouts without code changes
AI suggestion + human approval
Matching was checked against 238 scenarios and duplicate merging has 25 focused tests. Client names and volumes are omitted.
CASE / 03

Polish e-invoicing · Anonymized

Connecting KSeF to a legacy invoicing system

Before: the legacy PHP system could not communicate directly with Poland's National e-Invoice System (KSeF), which requires newer authentication and invoice formats.After: I added a separate Python service that converts invoice data to FA(3) XML, authenticates with KSeF, sends invoices, checks their status, and finds an invoice submitted earlier. The old system can now use KSeF without a risky rewrite of the whole application.

KSeFPythonFastAPIPHPFA(3) XMLAlready-submitted invoices

What now works

Legacy PHP → Python service → KSeF
Send invoices & check status
Find an already-submitted invoice
Implemented as four API endpoints with FA(3) XML generation and four focused tests. Business details remain private.
CASE / 04

Loyalty-card platform · Public source

A loyalty-card platform designed and built from scratch

Before: I designed and wrote the first version from scratch for one business. It handled customer registration, physical cards, Dotykačka customer synchronization, Wallet cards, and email, but all customers, settings, and integrations belonged to one setup.After: I then developed it into a Django platform that keeps every business separate. A customer scans a physical card; the system assigns it, records consent, synchronizes the customer with Dotykačka, prepares a personal Apple Wallet or Google Wallet card, updates Brevo, and sends the card by email. Each business has separate branding, inventory, users, and encrypted integration credentials. If an external service fails temporarily, background jobs can retry safely.

DjangoDotykačkaApple WalletGoogle WalletBrevoMariaDB

What now works

267 customer records preserved
Complete 600-card inventory preserved
228 automated tests · SQLite + MariaDB
The technical expansion into a platform is complete. Adding more businesses remains disabled until final human, provider, and printing checks. Public source is available; customer data, credentials, and generated Wallet cards stay private. View public source
01

ETHWarsaw 2024

CoinSwipe: 1st place in the DragonSwap bounty

CoinSwipe is a Telegram app that combines discovering memecoins, trading them, and meeting people interested in the same coins. I built it with my team during ETHWarsaw 2024; we won first place in the DragonSwap bounty, and 42 Warsaw later featured the project.

02

HumanTech Hack 2024 · Team Warsaw Moss

Eco Navigate: 3rd place

Before: route planners mainly compared time and distance.After: Eco Navigate compares walking and cycling routes through Warsaw using data about nearby trees, parks, and other green areas. It recommends a greener option while showing its distance, time, and extra detour, and it also displays live air-quality stations.

03 / About

Engineering guided by real business experience.

I work where software, servers, data, and day-to-day business operations have to work together.

My engineering work covers Python and PHP applications, APIs, SQL, Linux, Docker, production problems, and legacy modernization. Nine years of running my own e-commerce business taught me to judge technology by whether it saves time, reduces risk, and works reliably, not by whether it is new.

I am most useful when the starting point is unclear, several systems must exchange data, and someone needs to take responsibility from the first discussion through production support.

How I workDiscover → design → deliver → operate
Current focusCloud integrations and AI-assisted work with human approval

04 / Mentoring

Good questions create independent progress.

As a Space Eagle at 42 Warsaw, I mentored new participants during the four-week Piscine, an intensive, on-campus introduction to peer-to-peer and project-based learning. I helped them reason through problems and find their own next steps. My experience also includes mentoring at HackYeah 2026.

See my mentoring approach
Dmytro explaining code beside a laptop while his dog listens
Explaining code to my most honest reviewer.

05 / Toolkit

Tools I use to build, connect, and run systems.

Build

Python · PHP · JavaScript · FastAPI · Django · Flask

Integrate

REST APIs · Webhooks · SQL · ETL · CSV / JSON / XML

Operate

Linux · Docker · Proxmox · CI/CD · Logging · Backups

Data

PostgreSQL · MySQL · Redis · Validation · Data mapping

Expanding

AWS · Terraform · Kubernetes · Observability · AI evals

Business

Discovery · Architecture · E-commerce · Mentoring · Ownership

06 / Contact

Have a difficult system to connect?

Email copied