today's lesson:
Islam Kharfi
Biotechnology Engineer · Clinical AI & Bioinformatics
Biotechnology Engineering Graduate · Master's in Biotechnology and Health at ENSB, Algeria. I build AI systems at the intersection of biomedical data, signal processing, and bioinformatics.
Skills
Languages
- c
- html
- matlab
- Python
- TypeScript
Frameworks & Libraries
- AI
- deep learning
- image processing
- LLM fune tuning
- machine learning
- Next.js
- React
- signal processing
Projects
problem nº 1
BioPrep-AI↗
LLM-driven biosensor preprocessing: audit signal/image files, generate sandboxed pipelines, and containerize with offline-safe fallbacks.
solution sketch
BioPrep-AI
biosensor, codex, computer-vision, docker, llm, openai, preprocessing, python, signal-processing, streamlit
open on github ↗
problem nº 2
AffiBio↗
Gene and variant analysis tool combining Affinage (ClinVar/PubMed literature evidence) with Biomni (LLM-powered biomedical reasoning). Analyze genes, phenotypes, or VCF files via CLI or Streamlit web UI, with dynamic tool selection, selective datalake downloads, and structured agent output enforcement.
solution sketch
AffiBio
ai, bioinformatics, biomni, genomics, llm, python, streamlit, variant-annotation
open on github ↗
problem nº 3
NeuroPredAI↗
NeuroPredAI is an end-to-end multi-modal machine learning platform that classifies psychiatric disorders from 19-channel EEG data using a stacked EEGConformer, CatBoost, and SVM ensemble. Deployed across Hugging Face Spaces and Vercel, it features an integrated fine-tuned LLM that generates automated, plain-language explanations for diagnostic reports.
solution sketch
NeuroPredAI
ai, deep learning, machine learninh, llm, eeg, signal, preprocessing
open on github ↗
GitHub activity
173 contributions in the last year. Attendance: perfect
Contact
Office hours are open. Leave a note on the board and I'll get back to you.
