Journey - Career Growth & Learning Timeline

June 2026 Learning

Independent Projects & Reinforcement Learning

Shipped TLiquid, an open-source LLM menu-bar translator for macOS, started building a multiplayer browser game, and began a ground-up study of reinforcement learning — from MDPs and value functions to hands-on PyTorch.

Skills:

Reinforcement Learning
PyTorch
Rust
Tauri
TypeScript

What Changed

2026 became the year of building things I fully own. The first result is TLiquid — a tiny, open-source, bring-your-own-key LLM translator that lives in the macOS menu bar. Taking it from product spec to a live, downloadable app at tliquid.app meant owning every layer myself: Rust system integration, Svelte UI, release packaging, and a public roadmap. In parallel, I’m building a multiplayer browser game — still in development and under wraps for now, but it’s been a deep dive into real-time networking and game simulation.

The second thread is deliberate study: reinforcement learning from first principles. Rather than jumping straight to frameworks, I’m working up from the foundations — Markov decision processes, value functions, and the Bellman equations — alongside hands-on PyTorch fundamentals (autograd, building networks from scratch) and Gymnasium environments. Coming from years of production backend engineering, RL feels like the natural next frontier: it’s where the systems thinking I’ve built professionally meets the learning-based methods reshaping the field. The goal is depth over speed — understanding why the algorithms work before reaching for the libraries that hide them.

November 2024 Transition

Embracing AI-Driven Development

Joined a translation platform as a Senior Software Engineer, integrating large language models and AI-powered workflows into production systems using Langchain4j, Spring AI, and event-driven architecture.

Skills:

Langchain4j
Spring AI
Kafka
Kotlin

What Changed

Moving to a translation platform in late 2024 marked the beginning of an entirely new chapter. The platform combined traditional software engineering with cutting-edge AI capabilities, using Langchain4j to orchestrate LLM-powered translation and content processing pipelines. I found myself working across a wide version spectrum — from Java 8 legacy modules to Java 24 greenfield services — while also picking up Kotlin for new microservices where its concise syntax and coroutine support made it a natural fit.

The most exciting part of this role has been bridging the gap between AI capabilities and production-grade engineering. Building with Langchain4j and integrating n8n workflow automation taught me that AI in production is not just about calling a model API — it is about designing robust pipelines with proper error handling, fallback strategies, token budget management, and observability. Kafka serves as the backbone for our event-driven architecture, ensuring that translation jobs flow reliably through multiple processing stages. This role has reinforced my belief that the best engineers in the AI era are not the ones who can prompt a model, but the ones who can build the reliable, scalable systems around it.

September 2023 Learning

Leading Monolith Decomposition

Led the architectural effort to decompose a monolithic god-service into well-bounded microservices. This project was a masterclass in domain-driven design, API contracts, and managing distributed complexity.

Skills:

Microservices
Spring Boot
API Design
Domain-Driven Design

What I Learned

The monolith had grown into what the team affectionately called the “god-service” — a single deployable unit that handled everything from data ingestion and transformation to reporting and notifications. Deployments were nerve-wracking, test suites took over an hour, and a bug in one module could cascade into seemingly unrelated features. I took the lead on decomposing this monolith into a set of focused microservices, each aligned with a distinct business domain.

The process was as much about people and process as it was about technology. We started by mapping bounded contexts using domain-driven design principles, identifying natural seam lines where the monolith could be split without breaking existing functionality. We introduced API contracts between services, established shared libraries for cross-cutting concerns, and implemented the strangler fig pattern to migrate functionality incrementally rather than attempting a risky big-bang rewrite. The result was a system that was easier to deploy, test, and reason about — and a team that had a much clearer ownership model. It remains one of the most impactful technical initiatives I have led.

June 2022 Learning

Deep Dive into Medical Systems

Developed deep expertise in healthcare data systems, learning about data security requirements, regulatory compliance, and the critical importance of data integrity in medical contexts.

Skills:

Data Security
MySQL
Flyway
Legacy System Modernization

What I Learned

A year into the medical data platform, I had moved beyond just writing features and was deeply immersed in the unique challenges of healthcare technology. Working with sensitive patient data meant that every database migration managed through Flyway had to be meticulously planned and reversible. Every API endpoint required careful consideration of access controls, audit logging, and data anonymization. The consequences of a data breach or corruption in this domain were not just financial — they were deeply personal to the patients whose records we managed.

I also gained significant experience in legacy system modernization. The platform had evolved over years, accumulating technical debt in the form of tightly coupled components and inconsistent data models. I led efforts to introduce proper database versioning with Flyway, refactor critical data access layers, and establish security review processes for any changes touching patient data. This experience taught me that modernizing legacy systems is not about rewriting everything from scratch — it is about making careful, incremental improvements while keeping the system running reliably for the people who depend on it every day.

July 2021 Milestone

Promoted to Senior Java Developer

Stepped into a senior engineering role on a medical data platform, taking on responsibilities for system architecture, technical leadership, and mentoring. A recognition of three years of consistent growth.

Skills:

Java 17
MySQL
System Architecture
Technical Leadership

What Changed

Being promoted to Senior Java Developer and joining a medical data platform marked a significant inflection point in my career. For the first time, I was explicitly responsible not just for writing code, but for the technical direction of the systems I worked on. The medical domain brought its own gravity — the data we handled directly impacted patient care, so there was zero tolerance for carelessness.

I embraced Java 17 and its modern features like sealed classes, records, and enhanced pattern matching, which allowed us to write more expressive and type-safe code. But the biggest shift was in how I spent my time. Architecture discussions, pull request mentorship, incident post-mortems, and cross-team coordination became as important as coding. I learned that senior engineering is fundamentally about multiplying the effectiveness of the people around you, not just being the fastest individual contributor.

May 2020 Transition

Tackling Enterprise ERP Systems

Moved to a fashion industry ERP project, confronting the complexity of enterprise resource planning domains. This role pushed me to master complex business logic and rigorous testing practices.

Skills:

PostgreSQL
Spring MVC
JUnit 5
Cucumber

What Changed

Joining the fashion ERP project was a deliberate step outside my comfort zone. After two years on a media platform where the domain was relatively intuitive, I found myself navigating the intricate world of inventory management, order fulfillment, supply chain workflows, and pricing rules that varied by region, season, and customer tier. The business logic was dense, and getting it wrong meant real financial consequences.

This role transformed me into a more disciplined engineer. I leaned heavily into behavior-driven development with Cucumber, writing scenarios in close collaboration with domain experts to ensure every edge case was captured before code was written. PostgreSQL became my daily companion for complex queries involving multi-table joins across inventory, orders, and logistics data. I learned that in enterprise systems, understanding the domain is just as important as understanding the technology — and that the best engineers are the ones who can speak both languages fluently.

August 2019 Transition

Growing into a Middle Developer

Transitioned from a junior to a middle developer through consistent delivery and expanding technical ownership. Began taking on more complex system design tasks and mentoring newer team members.

Skills:

Java 11
Spring Boot
System Design

What Changed

After a year and a half of steady growth on the web-media platform, I reached a turning point where I was no longer just implementing tasks handed to me — I was actively shaping technical decisions. Migrating our codebase from Java 8 to Java 11 was one of the first initiatives I drove, navigating module system changes and identifying deprecated APIs that needed replacement across the project.

This transition was less about a title change and more about a shift in mindset. I started thinking beyond individual tickets: how does this feature affect the broader system? What are the operational implications? How can we make this easier to maintain six months from now? I began participating in code reviews not just as a reviewer, but as someone who could articulate architectural patterns and guide less experienced developers toward cleaner solutions. It was the period where I stopped thinking of myself as someone who writes code and started thinking of myself as someone who builds systems.

June 2019 Milestone

Master's Degree in Computer Science

Graduated with a Master's degree in Computer Science from Kharkiv National University of Radio Electronics. Balancing full-time work and academic research sharpened both my technical depth and discipline.

Skills:

Computer Science
Software Engineering
Algorithms

What Changed

Completing my Master’s degree while working full-time as a Java developer was one of the most demanding periods of my life, but also one of the most rewarding. At Kharkiv National University of Radio Electronics, I deepened my understanding of algorithms, data structures, distributed systems theory, and software engineering principles that I was already applying daily in my professional work.

The real value of the degree was not just the credential — it was the discipline of rigorous thinking. Writing my thesis forced me to approach problems methodically, review existing literature, and defend my design decisions with evidence. These habits carried directly into my engineering work, making me a more thoughtful developer who could reason about trade-offs rather than just shipping the first solution that compiled.

September 2018 Learning

First AWS Cloud Integration

Gained hands-on experience with AWS cloud services including S3, RDS, Lambda, and Transcribe while building features for the media platform. This was my introduction to cloud-native thinking.

Skills:

AWS
S3
Lambda
Transcribe

What I Learned

About seven months into my first role, I was tasked with integrating several AWS services into our media platform. We needed to store and process large media files using S3, run serverless processing jobs with Lambda, and leverage AWS Transcribe to generate transcriptions of audio and video content. It was my first real encounter with distributed cloud architecture, and it fundamentally changed how I thought about building software.

Working with AWS taught me that modern applications are rarely self-contained monoliths — they are compositions of managed services stitched together with careful API design and event-driven patterns. I learned to handle eventual consistency, manage IAM permissions, deal with cold starts in Lambda functions, and design resilient workflows that could recover from partial failures. These lessons in cloud-native development became a foundation I would rely on throughout the rest of my career.

February 2018 Milestone

Started My Java Career

Joined a web-media platform as a Junior Java Developer, marking the beginning of my professional software engineering journey. This was my first opportunity to work with enterprise-grade Java technologies in a production environment.

Skills:

Java 8
Spring Boot
Docker
REST APIs

What Changed

Landing my first role as a Junior Java Developer at a web-media platform was a pivotal moment. I went from academic projects and personal experiments to writing code that served real users at scale. The stack was built on Java 8 with Spring Boot, containerized with Docker, and deployed to AWS — a far cry from the simple console applications I had been writing in university.

The learning curve was steep but exhilarating. I quickly picked up how to design RESTful APIs, write meaningful unit tests, and navigate a production codebase with thousands of lines of code. Working alongside experienced engineers taught me not just how to write code, but how to think about software — considering maintainability, performance, and the people who would read my code long after I moved on.