| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Original HTTPS Page] |
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.
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abuseI build things that ship: Elexon BMRS forecasting pipelines, local-LLM phone systems, and production web platforms. BSc Computer Engineering, 1st in class (GPA 18.87/20). Enterprise Scholarship. Student Rep Gold Award. Currently wrapping up MSc Data Science at UWE Bristol.
| Project | What it does | Stack | Status |
|---|---|---|---|
| GB Balancing Mechanism Forecasting | Forecasting total GB balancing mechanism cost per settlement period, a target no published GB study models directly. LEAR baseline vs XGBoost and LSTM, rolling-window evaluation, Diebold-Mariano testing | Python · XGBoost · LSTM | 🔵 Due Jan 2027 |
| Reception System | Local LLM phone handler for SMEs, prototype stage: Whisper STT, fine-tuned LLaMA/Phi-3, Google Calendar API | Python · FastAPI · Ollama | 🟡 Prototype |
| SmartQuote Assistant | Email extraction + quote reply automation, 96% field accuracy | Python · Ollama · OpenAI | ✅ Done |
| Whisper STT Eval | STT evaluation on LibriSpeech: WER, CER, latency + live Streamlit dashboard | Python · Faster-Whisper | ✅ Done |
| ML Practice | One algorithm per week: theory notes + notebook, KNN to MLP | Python · scikit-learn | 🟢 Active |
| Crime & Deprivation | Crime vs deprivation across English police areas | MongoDB · PostgreSQL | ✅ Done |
| Retail KPI Pipeline | End-to-end BI pipeline over £10.6M sales dataset. ETL, MySQL, Streamlit | Python · Pandas · MySQL | ✅ Done |
| File Organizer | Python CLI that auto-sorts files by type, with dry-run, undo, custom config | Python | ✅ Done |
Languages
Data & ML
Currently learning
Databases & Infrastructure
Bristol, UK · Open to internships and graduate roles · knhnf.github.io
Predicting UK residential retrofit potential from EPC open data using Random Forest, XGBoost, Logistic Regression and SVM under nested cross-validation
Jupyter Notebook 1
A simple and customizable Python tool that automatically organizes files into folders, with interactive mode, undo support, dry-run preview, and Windows-safe output.
Python
Personal profile page — links to portfolio, CV, and projects.
HTML
Small BI-style data validation and reporting pipeline using Python and MySQL.
Jupyter Notebook
Email extraction and quote automation using local models and GPT. 96% field accuracy
Python
| Back | FazBrowse Home | New Git URL |