LLM Engineer (AI)

A high-growth Israeli startup operating at the intersection of Fintech and Health-Tech. The company has developed an intelligent platform designed to eliminate billions of dollars in fraudulent or erroneous medical claim billings for US-based enterprises.

By providing real-time monitoring and automated auditing of healthcare claims, the platform allows self-insured employers to significantly reduce operational costs.

Founded by a leading venture capital group and backed by substantial early-stage funding, the company is rapidly expanding its R&D capabilities.

Located in Central Tel Aviv (prime access to the train station), the company operates a hybrid work model (one day per week from home).

Role Overview:

• Joining a core engineering team working in close collaboration with Data Scientists to deliver AI-driven solutions.
• Developing and integrating Production-level AI Pipelines and LLM-based components into the core product.
• Implementing advanced techniques such as RAG (Retrieval-Augmented Generation), Agentic Workflows, and Prompt Engineering.
• Responsible for the end-to-end lifecycle of AI Services: from model integration and data enrichment to backend connectivity.
• Optimizing system Performance, Latency, and Accuracy within multi-cloud environments (AWS/GCP/Azure).
• Tackling complex challenges involving Model Optimization, Bias Detection, and Concept Drift in large-scale data environments.

Requirements:

• 5+ years of professional experience in Software Development – Mandatory.
• Proven experience in developing with LLM frameworks and RAG architectures – Mandatory.
• Strong proficiency in Python and modern Machine Learning frameworks – Mandatory.
• Hands-on experience with Cloud Infrastructure (AWS, GCP, or Azure) – Mandatory.
• Deep understanding of AI Infrastructure, Data Analysis, and Agentic Workflows.
• Ability to bridge the gap between model research and high-scale production engineering.
• Experience in the Fintech, Insurtech, or Healthcare sectors – A significant advantage.