AI Engineer

The core responsibilities for the job include the following:

Generative AI Core Engineering and Agent Orchestration:

• Help architects and deploy multi-agent workflows using modern orchestration frameworks (LangGraph, LlamaIndex, Bedrock, and Vertex AI) to solve complex, multi-step business problems.
• Design and build Retrieval-Augmented Generation (RAG) systems, fine-tuned LLMs, high-performance embedding pipelines, and document intelligence solutions.
• Contribute to productionizing LLM-based agents for claims validation, pre-authorization, sales advisory, care management, and internal AI copilots.
• Support work on advanced models, including transformers, diffusion models, and multimodal (vision, speech) models for specialized health-tech challenges.

Full-stack product engineering:

• Work across the entire UI API model workflow, ensuring seamless, high-performance integration from front-end to inference.
• Build user interfaces using modern frameworks (Next.js, React, Streamlit) for both customer-facing and internal operational tools.
• Develop scalable backend microservices (Python/FastAPI, Node.js/Express) optimized for LLM inference and large-scale data serving.

LLMOps and Deployment:

• Ship models to production using modern MLOps tooling (MLflow, CI/CD, Docker, and Kubernetes).
• Help optimize latency, throughput, inference cost, and security for deployed GenAI services.
• Build monitoring and observability for GenAI: model drift, hallucination and toxicity rates, guardrail/compliance violations, and agent decision tracing.

Requirements:

• Strong fundamentals in algorithms, data structures, distributed systems, and applied machine learning/deep learning.
• Hands-on exposure to LLMs and at least one agent/orchestration framework (LangChain/LangGraph, LlamaIndex, Bedrock, or Vertex) through projects, internships, or research.
• Working proficiency in Python; familiarity with a modern web stack (React/Next.js frontend, FastAPI or Node.js back-end).
• A builder's mindset is comfortable in ambiguous, 0 to 1 environments where you help define both the product and the architecture. Preferred experience with multi-modal models (vision, speech) relevant to healthcare/insurance documentation.
• Exposure to fine-tuning LLMs or training custom small models for domain-specific tasks.
• Interest in or exposure to healthcare, insurance, or financial services domains.
• Familiarity with optimizing LLM inference (quantization, compilation, specialized hardware).