Daniel Rotkopf
Full-stack engineer
I build web applications and the systems that run behind them. I prefer working across the whole of something rather than one slice of it.
I also work in RF, and on machine learning.
RF
I work in RF and signal processing. Taking a signal apart, working out what is actually inside it, and building the tooling around that.
Machine learning
What interests me is the part close to the data and the model itself. How a network is actually trained, why one architecture fits a problem better than another, and what to do when the data is messy or missing. I have worked with neural network fundamentals and backpropagation, sequence models for text, image classification, and using language models to label data.
Projects
VidCompiler
Free file storage, hidden inside YouTube videos
Turns any file into a video and hands back one link. Download it later and every byte comes back exactly as it was.
- Python
- C
- NumPy
- Flask
- FFmpeg
- +5
Reddit Stock Sentiment
Scraper, LLM labelling, LSTM classifier, Flask app
A sentiment classifier for Reddit stock discussion, including the dataset it trains on.
- Python
- TensorFlow/Keras
- LSTM
- GPT-2 tokenizer
- OpenAI API
- +3
Neural Network Library in NumPy
Forward and backward passes by hand, then an MRI classifier on top
A small deep learning library written in NumPy, and the brain scan classifier I built with it.
- Python
- NumPy
- h5py
- scikit-learn
- Pillow
- +1
Jarvis
Spotify playback bot
Started, not finished. Currently a repo with a plan in it.
- Python
About
Most of my work is full-stack web development. I build the interface, the API behind it, and the pipeline that gets it deployed, and I am comfortable owning all three rather than handing parts off.
On the front end that means React and TypeScript, with MUI for components. Behind it, Postgres for storage and Redis for caching. Everything runs in containers on Kubernetes.
Frontend
- React
- TypeScript
- JavaScript
- MUI
- HTML
- CSS
Backend and data
- Python
- Flask
- Postgres
- Redis
- SQLite
- REST APIs
Infrastructure
- Docker
- Kubernetes
- CI/CD
- Git
- Linux
- nginx
RF and signals
- RF
- Signal processing
- CRF
- Calculus
Machine learning
- TensorFlow/Keras
- NumPy
- scikit-learn
- Pandas
- Selenium
- OpenAI API
- Matplotlib