Dominik Czerwiński

Solution Architect | .NET | Distributed Systems | Cloud

I am a Solution Architect working primarily in the .NET ecosystem. I design business systems so they can evolve without accumulating unnecessary complexity - using DDD, modular architecture, and an approach matched to the scale and context of the system.

I am currently focused on building a practical AI-assisted software delivery model - AI as an integral part of the full development lifecycle.

What I do

System architecture

I design backend system architectures - from modular monoliths to distributed systems. I focus on clear boundaries, consistency and long-term evolution.

Requirements and system boundaries

I translate business requirements into domain boundaries, APIs, and component responsibilities - working with stakeholders before the design decisions are made.

.NET and Cloud

I build solutions using .NET and Azure, focusing on scalability, reliability and maintainability in production environments.

AI-assisted SDLC

I build practical AI-assisted software delivery models - AI integrated across the full development lifecycle, from specification through to production.

Integrations and APIs

I design and develop APIs and system integrations. I maintain contract consistency, stability and the ability to evolve without frequent breaking changes.

Team alignment

I help define responsibility boundaries between frontend, backend, APIs and integrations so teams work from a consistent model.

How I think about architecture

I choose technologies that fit the problem, not the ones that happen to be trending

I match the architecture to the scale of the problem - I avoid over-engineering where a simpler model is sufficient

I design systems with long-term evolution, cost of change, and maintainability in mind

I make conscious trade-offs instead of aiming for a theoretically perfect solution

How I work

Understanding the problem

I start by understanding the business problem, system constraints, and where the current solution stops scaling or becomes too costly to change.

Architectural decisions

I design based on actual constraints, requirements, and scale - not on diagram-driven assumptions. I name the consequences of architectural decisions explicitly.

Simplifying systems

I reduce unnecessary complexity, define clear responsibilities and focus on consistency. I design so that adding a new feature next year doesn't require significant rework of existing components.

Staying close to the implementation

I stay involved beyond the design phase. I conduct design reviews, support the team on difficult cases, and make sure architectural decisions are reflected in production code.

Responsibilities

•Designing backend, integration, and API architectures
•Owning architectural decisions from problem analysis through to production
•Translating business requirements into system boundaries, APIs, and component responsibilities
•Maintaining architectural consistency across teams and system layers
•Mentoring developers and technical leads on architecture, DDD, and system design

Get in touch

If you're looking for a Solution Architect or Tech Lead who takes ownership of technical direction - get in touch.

Experience

CSHARK

Mar 2023 – Present

Domain: sales and order processing, SAP-integrated enterprise platform for a global industrial company

Senior Software Engineer (Technical Leadership / Architecture)

Mar 2023 – Present

Leading technical direction for sales and order processing capabilities.

Designed architecture for order intake, order change and order management flows
Introduced modular structure aligned with business domains and process boundaries
Designed consistent APIs for internal, frontend and external integrations
Clarified responsibilities and flows between backend, frontend, integrations and source systems
Coordinated technical decisions across teams
Improved documentation, onboarding and long-term maintainability
Developing practical AI-assisted development workflows supporting analysis, specification, implementation and review

Sente IT Systems

Feb 2013 – Feb 2023

Domain: enterprise financial platform with ERP and CRM modules, covering product sales, financial settlements, accounting, and broader business operations

Lead Architect

Sep 2016 – Feb 2023

Led architecture of a large-scale financial system under legacy constraints, high domain complexity, and limited architectural capacity across the organization.

Owned architectural direction for the entire system
Led transformation from legacy database-driven system to distributed architecture
Designed migration strategy using microservices and strangler pattern
Introduced DDD, CQRS, and event-driven architecture
Built internal frameworks and engineering standards
Coordinated architecture across 60+ developers and multiple teams

.NET Developer

Feb 2013 – Aug 2016

Development of a financial system using .NET and Firebird.

Education

Wrocław University of Science and Technology

2009 – 2014

Master of Science, Computer Science

IT Systems Design

2013 – 2014

Grade: Very Good

Bachelor of Engineering, Computer Science

2009 – 2013

Grade: Excellent

Key Projects

Legacy System Transformation

Sente IT Systems

Legacy system based on database logic and Firebird, with scalability issues and increasing operational risk.

Approach

•Migration using the strangler pattern.
•Gradual decomposition of the monolith into microservices.
•Moving parts of the processing to real-time.
•Introducing DDD as the foundation for domain modeling.

Impact

•Reduced operational risk by decoupling new service logic from high-risk legacy database code.
•Gave teams the ability to extend and replace individual modules without affecting the rest of the system.
•Consistent module structure and integration patterns across a 60+ developer project.

Sales and Order Management Domain Redesign

CSHARK

System with fragmented business logic, inconsistent APIs, and growing integration needs.

Approach

•Domain-driven modularization and clarification of responsibilities.
•API redesign focused on consistency and external integrations.
•Separation of logic between backend, frontend, and source systems.
•Alignment of technical decisions across teams.

Impact

•New order processing flows can be built on established domain boundaries without redesigning existing ones.
•Stable API contracts reduced breaking changes for frontend and external integration teams.
•Clearer responsibility boundaries reduced cross-team coordination overhead per change.

AI-Assisted Software Delivery Flow (Pilot)

CSHARK

Building a practical process for using Generative AI in team work: from project knowledge and specification through implementation to review.

Approach

•Designed a structured SDLC workflow with defined AI roles: spec writer, implementer, reviewer, and decision support.
•Built role-based agent configurations and instruction sets for GitHub Copilot and Claude Code.
•Applied the workflow on a real commercial project - iterating on knowledge base organisation, spec structure, and review cycles.
•Documented the methodology as a public blog series (ongoing, EN/PL) and published an open-source reference implementation.

Impact

•Process in active daily use on a real commercial project.
•Open-source reference implementation at github.com/dczerwinskipl/ai-sdlc-from-scratch.
•Blog series documenting practical lessons, tooling decisions, and control mechanisms for AI-assisted software delivery.

Tech stack

.NET / C#

Distributed Systems

System Architecture

Azure Cloud

Architecture & Design

Microservices • Modular Monolith • Domain-Driven Design • CQRS • Event-Driven Architecture • Clean Architecture • Legacy Modernization • Event Sourcing • Strangler Pattern • C4 / UML

Backend & APIs

.NET / C# • ASP.NET Core • REST API Design • gRPC • Entity Framework • Aspire

Cloud & Infrastructure

Azure • Docker • Kubernetes / AKS • CI/CD • Azure DevOps • GitHub Actions

Messaging

RabbitMQ • Kafka • Azure Service Bus • NATS

Data

PostgreSQL • SQL Server • MongoDB • CosmosDB • EventStore • Firebird

Frontend

TypeScript • React • React Native • Next.js

AI

GitHub Copilot • AI‑assisted development workflows • Instruction design for AI coding tools

Projects and writing

Alongside commercial work, I maintain personal projects and write. I use them to validate architectural approaches, tooling, and workflows before applying similar solutions in commercial projects.

Open Source

nevo

A .NET testbed for experimenting with CQRS, DDD, and Event Sourcing. I validate architectural patterns here before applying them in production. Stack: .NET, EF Core, PostgreSQL

ai-sdlc-from-scratch

Reference implementation of an AI-assisted SDLC: knowledge gathering, spec, implementation, and review workflows built from scratch. Stack: TypeScript, Node.js

nevo-ui

Monorepo with React component system and experiments with AI-assisted development. Focus: frontend architecture, design system and developer experience. Stack: React, Next.js, TypeScript, Turborepo

Writing

From Vibe Coding to AI Workflow

Blog series on building an organized AI-assisted SDLC: knowledge gathering, specification, implementation, review, and keeping AI contribution in check. In Polish and English.