Mamon Alsalihy
Verified Expert in Engineering
Artificial Intelligence Developer
San Francisco, CA, United States
Toptal member since July 18, 2025
Mamon is a generative AI expert with 4+ years of experience delivering full-stack NLP solutions that drive tangible business outcomes. He has built domain-specific LLMs, search and QA systems, and custom MCP servers and agents for real-time AI pipelines. At Brain Technologies and Flip AI, his work supported product launches and funding rounds. With deep expertise in Python, Hugging Face, LangChain, and PyTorch, Mamon helps clients go from idea to production fast, with impact.
Portfolio
Experience
- Python - 9 years
- Natural Language Processing (NLP) - 5 years
- Hugging Face - 5 years
- Prompt Engineering - 5 years
- Fine-tuning - 5 years
- Model Deployment - 4 years
- Customer Success - 3 years
- AI Agents - 2 years
Preferred Environment
Slack, Python, PyTorch, Hugging Face, LangGraph, OpenAI, Prompt Engineering, Retrieval-augmented Generation (RAG)
The most amazing...
...thing I've done is ship two LLMs that reduced CVS's annual costs from $7 million to only $200,000 across 10,000 stores in the pharmacy department.
Work Experience
Founding NLP Engineer
Flip AI
- Led cost reduction efforts for CVS's pharmacy translation system. Fine-tuned language models, implemented model distillation, and optimized prompts. The annual translation costs decreased from $7 million to $200,000 across 10,000 stores.
- Built custom MCP servers to expose customer data from databases and APIs to open-source LLMs—enabling our AI agents to integrate with enterprise systems, speeding up development of data-driven applications and expanding product capabilities.
- Architected a RAG-based Q&A pipeline integrating advanced vector search (HNSW) with fine-tuned LLMs, boosting accuracy by 20%, halving response times, and significantly improving user satisfaction.
NLP Engineer
Brain Technologies
- Increased sales by 10% by building a production-ready LLM classifier to classify user intent, better matching products to user queries, and achieved 15% higher accuracy. Deployed to AWS for low-latency performance under heavy user load.
- Built a no-code, drag-and-drop AI platform enabling non-technical users to chain models—ChatGPT and Midjourney—into creative workflows, generating multipage content and ads 5x faster, which became a core product attracting major enterprise clients.
- Cut customer research time by 50%, boosted engagement by 20%, and significantly improved sales conversions after building an AI summarization tool using GPT-3 and LangChain to transform thousands of user reviews into concise, feature-rich summaries.
Experience
Scientific Question Answering
https://github.com/mamonalsalihy/mle_tech_assessment• A lightweight retriever model to return the top-k most similar chunks of abstracts based on the query and title.
• An abstractive question answering model to generate concise and precise answers from the retrieved chunks.
Education
Master's Degree in Computer Science
University of California, Santa Cruz - Santa Cruz, CA, USA
Bachelor's Degree in Computer Science
University of California, Santa Cruz - Santa Cruz, CA, USA
Skills
Libraries/APIs
PyTorch
Tools
Slack
Languages
Python
Frameworks
LangGraph, Streamlit
Paradigms
Synthetic Data Generation
Platforms
Visual Studio Code (VS Code)
Other
Hugging Face, Software Development, Natural Language Processing (NLP), Fine-tuning, Model Deployment, Customer Success, AI Agents, FastAPI, Retrieval-augmented Generation (RAG), Open-source LLMs, OpenAI, Prompt Engineering
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring