10+
Years shipping products
AI Product Engineer with 10+ years across frontend and full-stack product delivery. Today I work on real product systems, and outside work I keep building independent projects to deepen my AI practice.
10+
Years shipping products
2+
Years focused on AI
4
Featured builds
Work and independent projects

Profile
Product-minded AI engineering with public proof.
Proof, in public
Professional work and independent builds live next to each other here on purpose. Together they show the direction I am pushing toward: product-minded AI engineering with real constraints, clean execution, and public proof.

Fotelik
Extraction, scoring, vector search, and explainable AI inside one independent product.

4cast
Zero-to-one full-stack build with product scope, auth, admin tooling, QA, and AI-assisted delivery discipline.
Selected work
The strongest signal is not one perfect case study. It is the combination of shipped work, AI product systems, and independent builds that make the transition legible.

Extraction, scoring, vector search, and explainable AI inside one independent product.
AI-powered political matching product built around extraction, scoring, retrieval, and explainability.
Primary proof
2025 - now
AI image product work centered on monetization, prompting, streaming UX, media pipelines, and the delivery systems around them.
Primary proof
2022 - now
Senior frontend role focused on making a complex production platform easier to change, safer to ship, and less burdened by legacy paths.
Primary proof

Zero-to-one full-stack build with product scope, auth, admin tooling, QA, and AI-assisted delivery discipline.Zero-to-one nightlife discovery PWA built with disciplined scope, a real operations layer, and an AI-assisted delivery workflow.
Primary proof
Supporting proof
The portfolio should lead with projects, but the background still matters. This is the delivery history and structured learning that support the move toward AI product engineering.
Experience
Delivery history that supports the current direction: product work, refactors, AI systems, and the engineering judgment behind them.
Timeline
Common
Dec 2022 - now
Frontend architecture, refactor execution, quality systems, and selected AI product capabilities across Common and Waku.
Patio
Jun 2021 - Nov 2022
Startup-speed product delivery with strong ownership across implementation, UX detail, and shipping quality.
Brainhub
Dec 2018 - Jun 2021
International product delivery with full-stack execution, CI discipline, and close collaboration with product teams.
Grape Up
Sep 2016 - Oct 2017
Pixel-perfect React and Redux work with strong testing discipline, SDK implementation, and close collaboration with US design teams.
Unity-t
Dec 2015 - Aug 2016
Foundational frontend ownership across UI design, framework migration, and close REST API collaboration.
Credentials
Structured learning and recent formal context, presented as supporting archive rather than headline proof.
Ledger
Przeprogramowani
Advanced course work focused on AI-first MVP delivery, production-grade engineering systems, legacy modernization, and structured AI-assisted workflows.
BRAVE
Hands-on course work around LLM APIs, multimodality, AI integrations, vector search, and practical agent-building patterns.
Writing and ongoing builds
Three lighter signals around the shipped work: public writing, recurring synthesis, and the workflows that keep turning rough ideas into reusable systems.
Writing
A deliberately small editorial layer: long-form essays on applied AI, trust, and production delivery, selected to show judgment rather than volume.
AI Digest
Recurring synthesis from raw sources into a public artifact: technical signals, source links, and structured takeaways for AI engineering work.
Latest digest
Daily AI Eng Digest (2026-03-13)
Top 5 X posts on practical AI engineering tools, agent orchestration, observability, and frameworks for production systems. Focus on actionable insights for full-stack JS/TS engineers shipping AI products.
Browse the AI Digest archive
Open AI DigestOngoing builds
Support rail
The lighter layer around the shipped work: systems for writing, research, reflection, and repeatable decision support.
A workflow for turning rough ideas into ranked business opportunities with structured research and scoring.
A Notion-powered personal operating system with assistants for reflection, planning, finances, and daily decision support.
Contact
The strongest fit is usually around AI product engineering, frontend-heavy product work, and systems that need both technical depth and product judgment. If the work matches, reach out.