marian.szawelski/portfolio
FREN

available — permanent role

Marian
Szawelski.
AI & Data engineer.

Machine learning, generative AI, RAG, agents: I design complete AI systems and put them in the hands of real users. AI that serves, not AI that gets in the way.

selected work

  • 2026

    Tessan AI Marketing — content generation platform

    In-production AI marketing platform for Tessan. RAG on pgvector + orchestration of 9 AI models (GPT-4o, Imagen, Veo, HeyGen) generating text, image and video while preserving brand voice. Autopilot scheduling, human-in-the-loop, Meta publishing. Solo build.

  • 2026

    Post-bac program pipeline — DuckDB · dbt · Dagster

    A post-secondary program reference dataset built from two open Parcoursup (French university admissions) datasets. Entity resolution across both sources, layered dbt modeling with 28 automated tests, full orchestration via Dagster. Solo project built to go deeper on the data engineering stack.

  • 2025

    Building footprint segmentation — U-Net

    Semantic segmentation model on high-resolution satellite imagery. Preprocessing, supervised training, IoU/Dice evaluation. Reproducible pipeline, commented notebooks.

about

From prep school to ENSISA, with a final year of studies in Japan. Final-year internship as AI project lead at Tessan. Now looking for a first permanent role in France.

full journey

stack

// languages
  • Python
  • TypeScript
  • JavaScript
  • SQL
// web
  • React
  • Next.js
  • FastAPI
  • Node
  • Tailwind
// ml
  • PyTorch
  • TensorFlow
  • scikit-learn
  • OpenCV
  • Transformers
// data
  • ETL/ELT
  • dbt
  • Dagster
  • DuckDB
  • Pandas
// ops
  • Docker
  • Postgres
  • Linux
  • Vercel
  • GCP