Mohamed Helayhel
Ai Software Engineer @Stealth Startup
Signup · Get unlimited contacts
WORK HISTORY
Ai Software Engineer @Stealth Startup
Designed AI agentic workflows (LangGraph, LangChain) integrating LLMs with Point-of-Sale data,enabling SMBs to make faster pricing, inventory, and growth decisions• Created LLM-powered tool that generates complex SQL queries/analytics frameworks and translates them into natural language insights, reducing ad-hoc analysis time from ~30 min to under 2 min• Led user testing with multiple clients, iterating weekly on prompts, UX, and workflow templates• Built RAG system for economic and competitor analysis that drove pricing optimizations for multiple SMB clients.• Developed MVP in 3 weeks using open-source stack (LangChain/ChromaDB), later optimizing for scale• Slashed LLM costs by 40% and latency by 50% via caching + model routing (prioritizing smaller LLMs)• Implemented hallucination guardrails for financial insights, reducing errors by 75% via sequential checks
EDUCATION
American University of Beirut
Bachelor of Engineering - BE, Mechanical Engineering
UC Berkeley Extension
Certificate Program in Data Science, Mathematics and Computer Science
SKILLS
ABOUT MOHAMED HELAYHEL
Experienced scientist and innovator, passionate about AI and leveraging data to its fullest potential.I am currently focused on researching and building projects in large language models (LLMs) and retrieval-augmented generation (RAG) to explore their applications in information retrieval and AI-driven decision-making. My goal is to leverage my combined experience in data science, engineering, and AI to create innovative solutions that improve our quality of life.Previously, I worked as an R&D Engineer at Dexcom for 4 years, where I developed and optimized early-phase health sensing technologies (biosensors) for clinical applications. I led various aspects of the process development roadmap, from designing experiments (DOEs) to performing data analytics and algorithm development with Python and MATLAB. I also have co-invented multiple patents in the field of health sensing and sensor signal optimization.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.