Skip to content
View KNHNF's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report KNHNF

Block user

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
KNHNF/README.md

Karan Homayounfar

Typing

MSc Data Science at UWE Bristol, focused on GB energy markets, forecasting and time-series ML. BSc Computer Engineering, 1st in class (GPA 18.87/20). Enterprise Scholarship. Student Rep Gold Award. I build things that ship: Elexon BMRS forecasting pipelines, local-LLM phone systems, and production web platforms.


🚀 Projects

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
SME Reception LLM Local LLM phone handler for SMEs: QLoRA fine-tuning, FastAPI backend, Faster-Whisper STT Python · FastAPI · Ollama 🟡 Prototype
EPC Retrofit Potential (ML) Predicting UK residential retrofit potential from EPC open data: Random Forest, XGBoost, Logistic Regression and SVM under nested cross-validation Python · scikit-learn · XGBoost ✅ Done
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
Retail KPI Pipeline End-to-end BI pipeline over £10.6M sales dataset. ETL, MySQL, Streamlit Python · Pandas · MySQL ✅ Done
NCC Hackathon build, see repo for detail (see repo) ✅ Done
File Organizer Python CLI that auto-sorts files by type, with dry-run, undo, custom config Python ✅ Done

🛠 Stack

Languages

Python SQL JavaScript HTML/CSS

Data & ML

Pandas NumPy scikit-learn Matplotlib QLoRA fine-tuning spaCy Whisper XGBoost

Databases & Infrastructure

MySQL PostgreSQL MongoDB Supabase FastAPI Docker Cloudflare Render Vercel

Currently learning: LangChain, RAG, SHAP, GitHub Actions, n8n


Portfolio CV LinkedIn Email

Bristol, UK · Open to internships and graduate roles · knhnf.github.io

Pinned Loading

  1. epc-retrofit-potential-ml epc-retrofit-potential-ml Public

    Predicting UK residential retrofit potential from EPC open data using Random Forest, XGBoost, Logistic Regression and SVM under nested cross-validation

    Jupyter Notebook 1

  2. gb-energy-forecasting-reproductions gb-energy-forecasting-reproductions Public

    Reproducing GB balancing-market forecasting papers with BMRS data, feature audits and rolling evaluation.

    Python

  3. sme-reception-llm sme-reception-llm Public

    LLM pipeline for SME voice assistant - QLoRA fine-tuning, FastAPI backend, Faster-Whisper STT

    Python

  4. england-crime-deprivation-analysis england-crime-deprivation-analysis Public

    Distributed Spark pipeline linking crime and deprivation across England at LSOA level, with MongoDB storage and an ML hotspot-classification model.

    HTML

  5. whisper-stt-eval whisper-stt-eval Public

    Speech-to-text evaluation with Faster Whisper on LibriSpeech. Compares tiny vs small models and baseline vs improved decoding across WER, CER, and latency. Includes a Streamlit dashboard.

    Jupyter Notebook

  6. retail-sales-kpi-analysis retail-sales-kpi-analysis Public

    Small BI-style data validation and reporting pipeline using Python and MySQL.

    Jupyter Notebook