I will be your data science consultant for analytics and ml projects
Data Scientist and ML Engineer for data pipelines, models and APIs
About this Gig
You have data but you are not sure what to do with it, or you have a model that stopped making sense.
I am a data scientist with a PhD in engineering sciences. I have built pipelines that process millions of records a day and models that are used in production, not only in notebooks.
On this call we go through your problem: what the data actually contains, which questions it can answer, what to build first and what to skip. You leave with a written summary of the recommended approach, the tools I would use and a realistic estimate of the work.
Typical reasons people book this:
- a dataset that is messier than expected and needs a plan before any modelling
- a model or dashboard that exists but does not answer the business question
- an analytics project that needs a second technical opinion before you commit budget
I work in Python and SQL, with pandas, scikit-learn, XGBoost, PyTorch, PostgreSQL and BigQuery. If your problem is better solved with one SQL query than with machine learning, I will tell you that too.
Send me a short note about your data before the call so I can prepare.
Technology:
Jupyter Notebook
•
Pandas
•
MySQL
•
BigQuery
Expertise:
Prediction
•
Forecasting
•
Statistics
•
Time series analysis
Programming language:
Python
•
SQL
•
Java
FAQ
What do you need from me before the call?
A short description of the problem and, if you can share it, a sample of the data or its schema. Ten rows are enough to start; we do not need the full dataset.
Is the call only advice, or do you also write code?
The basic call is advice plus a written summary. For hands-on work take the Standard or Premium package, where I review your code and model and change them directly.
Do you sign an NDA?
Yes. I work with clinical and industrial data under confidentiality agreements, so I can sign yours before you share anything.
Which tools do you use?
Python with pandas, scikit-learn, XGBoost and PyTorch, SQL on PostgreSQL, MySQL or BigQuery, and FastAPI or Docker when the result has to run as a service.
