Muhammad Ali

I’m an AI/ML Engineer who builds AI systems end-to-end—from training and deploying LLM agents to extracting insights that drive real decisions. I’m comfortable across the full stack, including pipelines, models, dashboards, and production deployments that hold up under real use.

In my recent work, I built an AI-powered elderly health monitoring system that uses WhatsApp automation, LLaMA-based text extraction, and Gemini vision to track vitals and medicines. The results were measurable: improved medication adherence by 30%, stabilized vitals, reduced HbA1c from 8.3 to 7.1 in two months, and added escalation alerting to caretakers.

I also worked on systems that scale and stay dependable in production-like conditions. For example, Rankistan ranks active Pakistani GitHub developers by scanning 40,000+ profiles in hourly workflow batches, using weighted scoring and deploying Groq AI summarization behind a Cloudflare Worker with rate limiting, CORS, and caching.

Beyond core ML, I enjoy building practical, user-facing products—like the Mehngai Monitor, which scrapes official Lahore rate-list images, normalizes items with Gemini OCR/semantic canonicalization, and serves analytics trends through a fast web dashboard. I value clean pipelines, strong evaluation, and shipping features that users can rely on.