Technical Architect & AI System Builder
I design reliable systems, AI-assisted workflows and data platforms.
With 14+ years in software engineering, I combine architecture, hands-on backend/full-stack development and technical leadership to build scalable, maintainable and future-proof solutions.
14+ years in software engineering
~28B records / ~2TB data migration
High-load platforms and multitenant architecture
AI agents, automation and ML-driven search
SLO/SLI, incident command and reliability culture
2 released games on Google Play
01
Profile
I am an experienced Technical Lead and Technical Architect with a background in building and scaling high-load platforms, designing robust system architectures and leading cross-functional teams through complex projects.
My engineering focus is backend and full-stack development with PHP, Node.js, Go, Python and C#, plus databases, distributed systems, data pipelines and AI-assisted engineering workflows. I have delivered large-scale data migrations, ML-driven content search and recommendation services, AI/automation workflows and frameworks that improve release confidence and system reliability.
Beyond coding, I focus on architecture governance, reliability, compliance and incident management. I define SLO/SLI practices, act as Incident Commander during critical outages, mentor engineers and standardize engineering practices.
02
Expertise
Backend & platforms
AI systems & automation
Data, search & AI-adjacent systems
Leadership & delivery
Compliance & automation
Game development & product
03
Services I can provide
Focused, practical help for teams that need architecture, AI-enabled automation and reliable delivery — from discovery to production rollout.
AI workflow automation
Design and implementation of practical AI-assisted workflows: task routing, context preparation, draft generation, review gates and integration with existing tools.
Agentic development systems
Multi-agent operating models for software teams: source-of-truth knowledge base, runtime handoff, Git-based validation, reproducible commands and safe staging-to-production boundaries.
AI-ready knowledge bases
Structuring documentation, decision logs and project state so AI tools can use reliable context instead of scattered chats and tribal knowledge.
Search and recommendation systems
Backend architecture for content tagging, semantic/image similarity search, OpenSearch relevance evaluation and recommendation workflows.
Architecture audit and technical leadership
Assessment of platform boundaries, reliability risks, data flows, API contracts, release process, observability and team delivery model.
Data pipelines and automation
Production-grade data workflows, Airflow orchestration, migration planning, monitoring, retry policies and operational handover.
04
Selected case studies
Public-safe summaries of systems and automation work. Detailed internal implementation specifics are intentionally omitted.
Multitenant platform architecture
- Problem
- Support scalable onboarding and separation between tenants/environments without operational chaos.
- What I built
- Designed and implemented multitenant architecture patterns for a large-scale platform.
- Result
- Enabled scalable onboarding and isolated deployments.
Consent management service
- Problem
- Build reliable consent handling that supports compliance and seamless integration across product systems.
- What I built
- Led the design of consent management service architecture.
- Result
- Improved compliance readiness, scalability and integration consistency.
28B-record data migration
- Problem
- Move and restructure very large datasets into a scalable query model.
- What I built
- Built migration and restructuring pipelines with partitioning and indexing, orchestrated via Airflow.
- Result
- Reliable replication and scalable queries for ~28B records / ~2TB.
Airflow orchestration platform
- Problem
- Standardize and improve reliability of production data workflows.
- What I built
- Deployed Airflow as a company-wide orchestration tool and built production DAGs.
- Result
- Cross-system pipelines with retry, SLA and monitoring.
ML-driven content search and recommendations
- Problem
- Improve content discovery, search relevance and recommendation quality.
- What I built
- Built a content tagging system, image recommendation/search services and OpenSearch relevance evaluation framework.
- Result
- Foundation for ML-driven features and significantly better search quality.
OpenAPI CRUD framework
- Problem
- Standardize API design and reduce repetitive implementation work.
- What I built
- Formalized code style conventions and built automatic CRUD generation around OpenAPI.
- Result
- More consistent API design and improved developer experience.
Multi-agent architecture for autonomous development
- Problem
- Coordinate autonomous agents so they can work on real projects without losing context, duplicating decisions or bypassing review/deployment boundaries.
- What I built
- Designed a multi-agent operating model where Obsidian acts as the source of truth, project state is captured in tasks and decision logs, agents can continue runtime events, and code/deployment changes are validated through reproducible commands and Git history.
- Result
- A repeatable autonomous development workflow with persistent context, clearer handoffs and safer staging-to-production boundaries.
Released Google Play games
- Problem
- Build and release complete casual games, not just prototypes, with enough polish to be published publicly.
- What I built
- Developed and released two mobile games on Google Play: WoodBricks and Sudoku Secrets.
- Result
- Hands-on product experience across game design, implementation, release preparation and post-release iteration.
Adobe After Effects asset library
- Problem
- Improve creative production workflow and asset reuse.
- What I built
- Built an asset library plugin for Adobe After Effects with a supporting web UI.
- Result
- More structured creative asset management.
Playable ads automation
- Problem
- Reduce manual work and improve consistency in mobile advertising creative production.
- What I built
- Built systems for generating playable ads for mobile advertising platforms.
- Result
- Faster and more repeatable creative production workflows.
05
Experience
X-FLOW — Technical Lead
Architecture, AI-assisted workflows, release governance, multitenancy, consent management, data pipelines, ML-driven search/recommendation, SLO/SLI and incident command.
FX Back Office — Senior Web Developer
Team/project leadership, Forex CRM support, feature development and payment integrations.
VOLKA GAMES — Automatization Lead / Senior PHP Programmer
Workflow automation, creatives database, Adobe After Effects asset library plugin and playable ads generation.
Ortnec Services Ltd — Senior Full Stack Engineer
High-load system development, MVP creation/testing, monitoring and performance.
WebYard Studio — Lead Full Stack Developer
Led two web development teams; maintained 10+ loaded projects and dozens of websites.
IRKIPEDIA — Web Developer
WordPress websites and photo stock development on Kohana.
DOT.RU Web Studio — Web Developer
Company CMS websites and website statistics parser on Kohana.
IDEAL Web Studio — Junior Web Developer
HTML coding and studio CMS websites.
06
CV, education & languages
Education
- Moscow Open Institute — Bachelor of Engineering, Computer Science and Computer Engineering (2016–2020)
- SkillFactory — Specialist, Data Science (2019–2020)
- Irkutsk State Pedagogical College #1 — Specialist, Pedagogy of additional education (2006–2010)
Languages
- Russian — Native or bilingual
- English — Professional working
- Greek — Elementary
Certifications
- Certificates of completion of two levels of the portfolio investor course
- Certificate in Business and Marketing
07
Architecture notes
Short, structured notes on how I think about systems, autonomy and engineering delivery.
Architecture as an operating model
Good architecture is not only diagrams. It is a set of boundaries, ownership rules, quality gates and feedback loops that let teams ship changes safely.
AI-assisted delivery
AI agents are most useful when they work inside explicit context: source-of-truth documentation, task states, decision logs, review gates and reproducible commands.
Reliability by design
I prefer systems where observability, SLO/SLI thinking, incident response and postmortems are part of the design rather than emergency additions after launch.
08
Contact
Open to relevant discussions about technical leadership, architecture, AI systems and product engineering.
Phone is intentionally not published. Use LinkedIn, Telegram or email.