Farhad Ali Farhat

AI/ML engineer (BSCS, UMT) focused on building and deploying LLM-powered and agentic systems. I work across the full path from prototype to production: LangChain, LangGraph, CrewAI, n8n, FastAPI, Supabase, and cloud deployment with CI/CD.

What I’ve built

• WhatsApp ordering assistant: an LLM-powered assistant deployed on Azure App Service with GitHub Actions CI/CD. It runs live on the WhatsApp Cloud API with HMAC-verified webhooks, staff approval buttons, and a notification design that avoids paid template messages. It uses Groq/Gemini with fallback, so it keeps working if one provider fails.

• CV screening pipeline: a LangGraph workflow that scores CVs against a job description requirement by requirement, with weighted scoring and a three-tier shortlist. It ships as a CLI and a Streamlit app.

• Published research: co-authored “FauxShield”, an explainable deepfake detection platform, published at IEEE FiCloud 2026 (Granada, Spain).

Experience

• AI/ML intern at Ebryx (current): completed structured roadmaps in classical ML, deep learning, and NLP, each with its own GitHub repo.

• AI/ML intern at PITB (2024).

Interests
Agentic workflows, LLM evaluation, and automation systems that hold up in production, meaning reliable, cost-aware, and observable. I’m looking for remote AI engineering roles where I can build real products and keep growing quickly.