Skip to content
View MohsenBah's full-sized avatar
🤓
"There's always 'one more thing' to learn."
🤓
"There's always 'one more thing' to learn."
6 followers · 12 following
  • Waterloo, ON

Block or report MohsenBah

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.

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
mohsenbah/README.md

Hi, I'm Mohsen 👋

I build AI systems, secure data pipelines, and self-hosted infrastructure.

Currently focused on:

  • 🧠 LLM systems: RAG, local inference, multimodal AI
  • 🔐 Privacy & security: PHI redaction, audit logging, safe AI
  • ⚙️ Backend & infrastructure: FastAPI, Docker, Proxmox, Linux networking
  • 🌐 Full-stack apps: React, TypeScript, API-driven systems

🚧 Current Focus

Secure Clinical AI Inference Pipeline

An end-to-end system for querying synthetic patient records with a local LLM while enforcing privacy and security controls.

Planned components:

  • FHIR-based synthetic data (OpenEMR + Synthea)
  • PHI detection & redaction (Microsoft Presidio)
  • Vector search (Qdrant / Chroma)
  • Local LLM inference (Ollama / llama.cpp)
  • FastAPI gateway with:
    • prompt injection protection
    • output filtering
    • rate limiting
    • audit logging

🧪 Other Work

  • Self-hosted AI + security lab (Proxmox, VLANs, DNS, monitoring)
  • Backend services with FastAPI (auth, logging, system design)
  • Frontend apps with React + TypeScript (Refine, MUI)
  • Earlier data science and ML projects

🛠 Tech Stack

Frontend

  • React, TypeScript
  • MUI (Material UI), Refine
  • Vite

Backend

  • Python, FastAPI, Flask
  • REST APIs, authentication, rate limiting

AI / ML

  • LLMs (Ollama, llama.cpp)
  • Embeddings, vector databases
  • PyTorch, NLP

Data

  • SQL, MySQL, Microsoft SQL Server
  • Pandas, NumPy, R

Infrastructure

  • Linux, Docker, Proxmox
  • Networking, VLANs, DNS
  • Git, Bash

📌 Featured Projects

Coming soon:

  • Secure Clinical AI Inference Pipeline
  • Self-hosted AI + Security Lab
  • FastAPI Security Gateway

Older repositories reflect my earlier work in data science and analytics.

📫 Contact

LinkedIn Medium Telegram

Pinned Loading

  1. Hopfield-Transformer-Hawkes-Process Hopfield-Transformer-Hawkes-Process Public

    Modeling Multivariate Hopfield-Transformer Hawkes Process: Application to Sovereign Credit Default Swaps

    Jupyter Notebook 5 2

  2. MedSecLab MedSecLab Public

    Reference architecture for securely deploying clinical AI systems with LLM security, SOC monitoring, and adversarial testing.

  3. clinical-ai-detections clinical-ai-detections Public

    Wazuh detection rules and security monitoring content for clinical LLM and healthcare AI workloads.

  4. clinical-ai-gateway clinical-ai-gateway Public

    Security-hardened clinical AI gateway with PHI redaction, prompt injection defense, audit logging, and local LLM inference.

    Python

  5. COVID19-news COVID19-news Public

    Jupyter Notebook