Profile picture

Product engineer shipping AI-native software end-to-end. Previously early engineer at YC-backed Scape (S24) and Legora (W24, now $5.6B valuation). 4× HackYeah hackathon participant, 1st place 2025.

Professional Experience
Product Engineer, TechTree⁠
  • •Built product from scratch for 1k freelance recruiters, driving 25% of our revenue
  • •Rebuilt job board from scratch - now sitting on 40k monthly visitors, 2k candidates applying monthly
  • 12/2025 – presentWrocław (remote)
  • •Built natural language search on 1 billion linkedin profiles
  • Software Developer, Scape (YC S24) - via AV⁠
  • •Early engineer at a Y Combinator S24 AI-native CRM startup
  • •Built the 2nd iteration of a Granola-like desktop meeting recorder (Electron, real-time transcription, summarization).
  • 02/2025 – 12/2025Remote
  • •Designed and shipped an app-wide AI agent giving users a single natural-language interface to drive the entire app.
  • •Led migration from a package-based monorepo to a monolithic architecture; resolved long-standing Postgres RLS performance bottlenecks.
  • Software Developer (via Alergeek Ventures), Legora (YC W24) - via AV⁠
  • •Early engineer through their Y Combinator W24 batch and $25M Series A (led by Redpoint; July 2024); legal-AI platform now at $5.6B valuation and $100M+ ARR (Series D led by Accel; backers include NVIDIA, Atlassian, Salesforce Ventures).
  • •Scaled file processing for user uploads and public document corpora (>1M docs per ingest) with Dagster, Redis and Node.js.
  • 02/2024 – 12/2024Remote
  • •Introduced an Nx-based monorepo, cutting build times and unblocking parallel team velocity.
  • •Migrated API layer from Next.js routes to Express.js; hardened auth, rate-limiting and input validation.
  • Junior Frontend Developer, Alergeek Ventures (AV)⁠
  • •Built a serverless game backend supporting 50,000 concurrent users.
  • •Cut CI pipeline time by 50% (≈5 min/run) via parallelization and caching.
  • 06/2022 – 02/2024Remote
  • •Frontend across several early-stage startup engagements.
  • Education
  • •GPA: 4.88 · final degree grade: 5 (Polish scale)
  • •Rector's Scholarship (Stypendium Rektora) - awarded multiple times for top academic performance; ranked #1 in the most recent cohort.
  • 10/2022 – 01/2026Wrocław
  • •Thesis: AI Present Finder - agentic gift-recommendation system, 6 specialized agents in event-driven microservice architecture.
  • 1 / 2
    Awards
    HackYeah 2025 — 1st place (15,000 PLN), Crowd-sourced multimodal transit search for rural Poland; ML on real-time road conditions, no carrier integration required.
    Google DeepMind × AI Tinkerers Warsaw 2026 — Finalist ($5k GCP), Agentic cooking assistant on Gemini 3.1 Flash; turned MacBook touchpad into kitchen scale and ordered missing ingredients via Wolt from a fridge photo.
    HackYeah 2023 — Finalist, Natural-language → SQL with open-source LLMs (Llama).
    HackYeah 2022 — 2nd place (10,000 PLN), AI-generated quizzes on top of the Polish Institute of National Remembrance archives.
    Projects
    discord-speech-recognition⁠, Open-source · 63★ on GitHub
  • •Extension for discord.js that makes voice bots as easy as text bots.
  • •9,000+ downloads, used in 100+ public OSS projects on GitHub.
  • •Stack: TypeScript, Jest, Google Speech-to-Text API; CI via GitHub Actions.
  • biedronka-client⁠, Open-source · 13★ on GitHub
  • •Reverse-engineered TypeScript client for the Polish Biedronka mobile app.
  • •Used by hobbyists to integrate loyalty data into personal automations.
  • AI Present Finder (B.Eng thesis)⁠, Multi-agent system · TypeScript
  • •6 specialized AI agents orchestrated in an event-driven microservice architecture.
  • •End-to-end agentic gift recommendation: prompt → reasoning → product retrieval → ranking.
  • Skills
    TypeScript

    Daily driver across all engagements; advanced generics & type-level features.

    React / Next.js

    SSR, server components, modern app router; built UIs serving 25k+ MAU.

    Dagster

    Pipelines processing >1M public docs at Legora.

    Docker / GitHub Actions

    Self-hosted infra, multi-stage builds, CI parallelization.

    Python

    Backend services and data tooling at TechTree; ML/LLM integrations.

    Node.js / Express

    Production APIs, auth, rate-limiting; migrated Legora from Next API routes.

    Postgres / Supabase

    Schema design, RLS, query/perf tuning at TechTree and Scape.

    Agentic systems · LLMs

    Gemini, OpenAI, MCP servers, Claude Code workflows.

    Languages
    English — C1
    Polish — Native
    Consent to data processing

    I hereby give consent for my personal data included in the application to be processed for the purposes of the recruitment process in accordance with Art. 6 paragraph 1 letter a of the Regulation of the European Parliament and of the Council (EU) 2016/679 of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation).

    2 / 2