Hi, nice to meet you! This is my personal website on which I explore Artificial Intelligence (AI), Mathematics, Music, Investing and Quantum Computing.
- I'm currently working as a freelance AI engineer and data scientist. Feel free to reach out on [LinkedIn](https://www.linkedin.com/in/carlo-lepelaars/) or [X](https://x.com/carlolepelaars) if you are interested in working together.
- I love reading and have compiled a list of my readings [here](/books).
- I play drums, guitar and love [sim-racing](https://en.wikipedia.org/wiki/Sim_racing)!

Past Work
CrowdCent (2022-2025)
I worked for [CrowdCent](https://crowdcent.com) as a (founding) data scientist on financial machine learning (ML), large language models (LLMs) and MLOps. CrowdCent is a startup building investment strategies driven by AI trained on investment community data. CrowdCent is collaborating with investment community [SumZero](https://sumzero.com) on a fund and also runs a fund around the financial data science tournament [Numerai](https://numer.ai).
DeepNewz (2024)
-
Large Language Model (LLMs) workflows
- Prompt Engineering
- Automated News Story Writing
- Topic Classification
- Language Translation
-
Prediction Markets
- Prediction Market Generation
- Prediction Market Quality Scoring
VBTI (2020-2021)
Computer Vision for agriculture:
- Rotational Estimation
- Object Detection
- Segmentation
- Keypoint Detection
Weights & Biases (2020-2021)
Wrote technical blog posts for Weights & Biases on a freelance basis:
Open Source Projects
numerblox
scikit-learn components for [Numerai](https://numer.ai). I built NumerBlox while at CrowdCent. NumerBlox is used within CrowdCent as part of its Numerai infrastructure.
blackscholes
Black-Scholes (financial option pricing) calculator. A lightweight package to determine option prices with [extensive documentation](https://carlolepelaars.github.io/blackscholes/3.the_greeks_blackscholes) on the math behind the Black-Scholes model. I also built a [tool that uses this library](/tools/blackscholes) under the hood.
auditus
Simple Audio Embedding Toolkit
q4p
Quantum computing for (Python) Programmers. This course makes learning Quantum Computing accessible for software engineers and aims to simplify quantum computing to its essence. It is built in public and new entries are gradually added.
skq
Scientific toolkit for Quantum Computing. Used in the q4p course.
fh-plotly
[Plotly](https://plotly.com) to [FastHTML](https://www.fastht.ml) conversion plugin.
flocky
Simple SDK for the decentralized AI platform [Flock.io](https://www.flock.io/).
ghostnet_tf2
At the time one of the first implementations of the [GhostNet](https://arxiv.org/abs/1911.11907) CNN architecture for Tensorflow 2.0.
Open Source Contributions
scikit-lego
Extra blocks for scikit-learn pipelines. Contributions include building support for multi-output models, custom arguments and scikit-learn compatibility.
FastHTML-Gallery
A gallery of [FastHTML](https://fastht.ml) examples. I contributed a variety of examples, including [Visualizations](https://fastht.ml/gallery/split_view?category=visualizations&project=bloch_sphere), [Image Uploads](https://fastht.ml/gallery/split_view?category=dynamic_user_interface_%28htmx%29&project=multi_image_upload), [Text and button animations](https://fastht.ml/gallery/split_view?category=dynamic_user_interface_%28htmx%29&project=animations), [Dynamic table loading](https://fastht.ml/gallery/split_view?category=dynamic_user_interface_%28htmx%29&project=click_to_load), [Audio player](https://fastht.ml/gallery/split_view?category=widgets&project=audio) and [PDF viewer](https://fastht.ml/gallery/split_view?category=widgets&project=pdf).
narwhals
A compatibility layer between DataFrame libraries (Polars, Pandas, PyArrow, Dask, etc.). I contributed the `skew` function and documentation.
fastcore
Supercharged functionality for Python programming. I wrote a blog post on [my top 10 favorite fastcore features](/posts/fastcore-10). I also made a small contribution to network functionality.
umap-learn
UMAP Dimensionality Reduction. I helped out with scikit-learn compatibility.
embetter
Embedding models easily accessible as scikit-learn components. I contributed bug fixes.
numerai-tools
Tools and utilities for Numerai. I helped out with metrics and dependency management. [NumerBlox](https://github.com/crowdcent/numerblox) uses numerai-tools to ensure consistency with how Numerai calculates payouts.