Skip to content
View ledostxx's full-sized avatar
1 follower · 2 following

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Highlights

  • Pro

Block or report ledostxx

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ledostxx/README.md

Hi, I'm Leonard

Data scientist finishing an MSc in Applied Information and Data Science at HSLU (Feb 2027). At Seibert Group I work in Revenue Operations, building churn and win prediction models, LLM classification pipelines with guardrails, and data automation on HubSpot, BigQuery and GCP.

Looking for: Data Science / ML roles in Switzerland from February 2027.

Selected projects

  • Hybrid Music Recommender – listening prediction on 7.5M Deezer events with affinity features, matrix factorization and LightGBM; 0.745 test AUC (team project)
  • FacADe – data lake and PostGIS warehouse on AWS estimating facade advertising exposure across Zürich (team project)
  • Swiss Boat Valuation – web scraping and hedonic pricing model for used motorboats; median error 24% (team project)
  • DB Delay Analysis – ELT pipeline on the Deutsche Bahn API and MongoDB Atlas; delay recovery and hub analysis across 20 stations

Stack

Languages: Python, SQL, R

ML: scikit-learn, LightGBM, XGBoost, pandas

Data & cloud: BigQuery, GCP, AWS, PostgreSQL/PostGIS, MongoDB, Docker

BI & automation: Tableau, n8n, LLM pipelines

leonarddost.de · LinkedIn

Pinned Loading

  1. deezer-hybrid-recommender deezer-hybrid-recommender Public

    Hybrid music recommender on 7.5M Deezer listening events: user affinity features, matrix factorization and LightGBM with chronological validation (0.745 test AUC)

    Jupyter Notebook

  2. facADe facADe Public

    Forked from fcampell/facADe

    City-scale data lake and PostGIS warehouse on AWS estimating facade advertising exposure across Zürich from open mobility data (traffic, pedestrians, bikes, public transport)

    Jupyter Notebook

  3. boat-valuation boat-valuation Public

    Hedonic pricing model for used motorboats in Switzerland: web scraping of boot24.ch, data cleaning pipeline and log-linear OLS regression

    Jupyter Notebook

  4. db-railway-analysis db-railway-analysis Public

    ELT pipeline on the Deutsche Bahn Timetables API and MongoDB Atlas: delay recovery, hub connectivity and peak-hour delays across 20 German main stations

    Jupyter Notebook

  5. Swiss-Publictransport-Winterbottlenecks Swiss-Publictransport-Winterbottlenecks Public

    R analysis of Zentralbahn delays in November 2024: delay propagation, weather effects and station recovery with LM, GAM, GLMs, SVM and a neural network

    HTML