22.03.2026 aktualisiert

**** ******** ****
100 % verfügbar

Data Science / Engineering & NLP

Berlin, Deutschland
Deutschland
BSc. Cognitive Science, Diploma in Mathematics
Berlin, Deutschland
Deutschland
BSc. Cognitive Science, Diploma in Mathematics

Profilanlagen

cv-christoph-schock.pdf
cv-christoph-schock-freelance.pdf

Über mich

• Experienced in both classical methods and modern machine learning • Value smart algorithms and effective representations of information • Focused on measurable results through robust baselines and sound metrics • Pragmatic approach: optimize where it matters, simplify where it doesn’t

Skills

Apache AirflowAlgorithmusGoogle BigQueryCollaborative Planning Forecasting And ReplenishmentDatenvisualisierungPythonMachine LearningNumPySQLZeitreihenanalyseMetrikenData ScienceLarge Language ModelsScikit-learnXgboost
* I’m interested in smart algorithms and good representations of information
* I have an excellent overview of classical algorithms and machine learning
* Baselines and metrics are important to understand the problems at hand
* I like to optimize where appropriate and figure out the right tool for a task
* 10 years of experience working with data in the Python ecosystem
* strong background in NLP / BI / SQL

Erfahrung

Berufserfahrung5 - 10 Jahre
Freelancer seit2017

Sprachen

DeutschMutterspracheEnglischverhandlungssicherSpanischgut

Projekthistorie

Product Recommendations

Konsumgüter und Handel

50-250 Mitarbeiter

Develop different product recommendation services for e-commerce retailer and optimize conversion via a/b tests.

Demand Forecasting for Warehouse Replenishment

Internet und Informationstechnologie

50-250 Mitarbeiter

Inventory Replenishment / eCommerce Retail (Data Science & Engineering, 18 months)
– Development of inventory replenishment tool (Python, Airflow, Flask)
– Demand forecasting (long-term, product-level)
– Training of junior colleagues

Media Content Recommender

Medien und Verlage

1000-5000 Mitarbeiter

Support development of content recommender using LLM embeddings

Data Pipelines for BI

Internet und Informationstechnologie

250-500 Mitarbeiter

– Build data pipelines with AWS Managed Airflow, Redshift, Python

Data Science & NLP prototyping for Railway Technology

Energie, Wasser und Umwelt

>10.000 Mitarbeiter

– Predictive modelling of CRM data (pandas, timeseries models)
– Build semantic search prototype (gradio, fasttext, numpy)
– Mechanical Vibration Analysis (scipy, numpy, altair)

Algorithm Development for OCR product

Energie, Wasser und Umwelt

10-50 Mitarbeiter

– Extract structured data from scanned documents (Google Document AI, scipy)
– Evaluate PoC performance, review and translate into production code

Trend Analysis

Konsumgüter und Handel

50-250 Mitarbeiter

2020 Build Trend Analysis Pipeline / Food & Beverages (Data Science, 3 months)
– Detect novel or emerging terms (n-grams) in social media streams
– Build visualisation to explore trend candidates

Build GCP Datalake

Konsumgüter und Handel

500-1000 Mitarbeiter

– Design architecture (Python, Airflow, BigQuery, kubernetes, Spark, GCP)
– ETL pipelines (SQL/Python) for sales, search rankings, product from diverse sources
– Establish best practices with BI manager and train developers
1/4– Terraform for GCP/kubernetes devops tasks

Develop Semantic Search Solution

Öffentlicher Dienst

250-500 Mitarbeiter

– Complement fulltext search of Apache Solr with neural language models
– Python REST API, scikit-learn, numpy, word2vec, fasttext
– Semantic search, similarity search, highlighting of relevant text snippets
– Cluster search results / classify and tag documents / find duplicates

Kontaktanfrage

Einloggen & anfragen.

Das Kontaktformular ist nur für eingeloggte Nutzer verfügbar.

RegistrierenAnmelden