Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m Vahit Uzunlar, an AI Architect & Product Owner with a proven track record of turning AI concepts into production-ready products. I specialize in end-to-end AI product development, contextual engineering, agentic design, and prompt optimization, delivering state-of-the-art performance (for example, a 4.1 nano model scored 68% initially, then rose to 85% and achieved a global top-5 ranking when optimized within the LM-XAI framework). My work spans health care, education, and nutrition, focusing on domain-adapted RAG pipelines and modular agent architectures that scale. I’ve led end-to-end design and delivery across multiple production-ready products under the LM-XAI umbrella, and I’ve built AI assistants for clients like Savion Clinic and Savion Patient, enabling both patient-facing and clinic-facing solutions to grow independently with a shared AI design language.…I’m Vahit Uzunlar, an AI Architect & Product Owner with a proven track record of turning AI concepts into production-ready products. I specialize in end-to-end AI product development, contextual engineering, agentic design, and prompt optimization, delivering state-of-the-art performance (for example, a 4.1 nano model scored 68% initially, then rose to 85% and achieved a global top-5 ranking when optimized within the LM-XAI framework). My work spans health care, education, and nutrition, focusing on domain-adapted RAG pipelines and modular agent architectures that scale. I’ve led end-to-end design and delivery across multiple production-ready products under the LM-XAI umbrella, and I’ve built AI assistants for clients like Savion Clinic and Savion Patient, enabling both patient-facing and clinic-facing solutions to grow independently with a shared AI design language.

Vahit Uzunlar

Architect, Product Designer, Data Scientist, +3





I’m Vahit Uzunlar, an AI Architect & Product Owner with a proven track record of turning AI concepts into production-ready products. I specialize in end-to-end AI product development, contextual engineering, agentic design, and prompt optimization, delivering state-of-the-art performance (for example, a 4.1 nano model scored 68% initially, then rose to 85% and achieved a global top-5 ranking when optimized within the LM-XAI framework). My work spans health care, education, and nutrition, focusing on domain-adapted RAG pipelines and modular agent architectures that scale. I’ve led end-to-end design and delivery across multiple production-ready products under the LM-XAI umbrella, and I’ve built AI assistants for clients like Savion Clinic and Savion Patient, enabling both patient-facing and clinic-facing solutions to grow independently with a shared AI design language.…I’m Vahit Uzunlar, an AI Architect & Product Owner with a proven track record of turning AI concepts into production-ready products. I specialize in end-to-end AI product development, contextual engineering, agentic design, and prompt optimization, delivering state-of-the-art performance (for example, a 4.1 nano model scored 68% initially, then rose to 85% and achieved a global top-5 ranking when optimized within the LM-XAI framework). My work spans health care, education, and nutrition, focusing on domain-adapted RAG pipelines and modular agent architectures that scale. I’ve led end-to-end design and delivery across multiple production-ready products under the LM-XAI umbrella, and I’ve built AI assistants for clients like Savion Clinic and Savion Patient, enabling both patient-facing and clinic-facing solutions to grow independently with a shared AI design language.

Available to hire

I’m Vahit Uzunlar, an AI Architect & Product Owner with a proven track record of turning AI concepts into production-ready products. I specialize in end-to-end AI product development, contextual engineering, agentic design, and prompt optimization, delivering state-of-the-art performance (for example, a 4.1 nano model scored 68% initially, then rose to 85% and achieved a global top-5 ranking when optimized within the LM-XAI framework).

My work spans health care, education, and nutrition, focusing on domain-adapted RAG pipelines and modular agent architectures that scale. I’ve led end-to-end design and delivery across multiple production-ready products under the LM-XAI umbrella, and I’ve built AI assistants for clients like Savion Clinic and Savion Patient, enabling both patient-facing and clinic-facing solutions to grow independently with a shared AI design language.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Language

English

Advanced

Work Experience

AI Architect & Product Owner at LMX AI

October 1, 2023 - Present

Designed and delivered end-to-end AI products under the LMX AI umbrella, including Savion Patient (end-user AI nutritionist with real-time meal recommendations, adaptive coaching, and goal tracking) and Savion Clinic (nutritionist-facing AI assistant automating clinic workflows from intake to reporting and follow-ups). Built from scratch with Django backends, Redis for state management, MongoDB for data/analytics, and modular agent architectures orchestrated via LangGraph and MCP. Created domain-adapted LKAi Student Assistant Platform for schools, delivering an intelligent conversational assistant via LangChain and custom agents; deployed on Elasticsearch, Redis, and MongoDB. Also developed a medical voice-to-text assistant (Whisper) to structure clinician notes, improving UX and documentation speed.

ML Developer in Test at Clinerion

November 1, 2022 - October 4, 2025

Worked on an end-to-end ML research project predicting Chronic Lymphocytic Leukemia (CLL)-like illness in collaboration with the University of Zurich and Bern University of Applied Sciences. Delivered full-cycle testing for an integrated ML system covering data ingestion, job scheduling, pipeline orchestration, and model inference. Performed end-to-end validation ensuring accuracy and reliability of predictions on real-world medical data. Installed and maintained complex test environments to validate ML pipelines and data access APIs; conducted hands-on testing of Airflow pipelines, model outputs, and data APIs.

AI Architect & Product Owner at Savion Clinic & Savion Patient

October 1, 2023 - Present

Designed and delivered two end-to-end AI nutrition assistant products built from scratch, with patient-facing Savion Patient providing real-time meal recommendations and adaptive nutrition coaching; Savion Clinic automates clinical workflows for nutritionists, from patient intake to policy generation, optimization, reporting, and follow-ups. Led backend in Django, Redis for state management, MongoDB for data and analytics, and modular agent architectures orchestrated via LangGraph and MCP, enabling scalable, cross-functional AI design language across patient-facing and clinic-facing contexts.

ML Developer in Test at Clinerion

November 30, 2022 - October 4, 2025

Worked on an end-to-end ML research project predicting Chronic Lymphocytic Leukemia (CLL)-like lymphoid patterns, in collaboration with University of Zurich and Bern University of Applied Sciences. Delivered full-cycle testing for an integrated ML system covering data ingestion, job scheduling, pipeline orchestration, and model inference; performed end-to-end validation ensuring accuracy and reliability; installed and maintained complex test environments for validating ML pipelines and data access APIs. Tools: Python, PyTorch, TensorFlow, Apache Airflow, BentoML, Bash, R scripting.