Senior AI Engineer

Inovalon was founded in 1998 on the belief that technology, and data specifically, would empower the transformation of the entire healthcare ecosystem for the better, improving both outcomes and economics. At Inovalon, we believe that when our customers are successful in their missions, healthcare improves. Therefore, we focus on empowering them with data-driven solutions. And the momentum is building.

Together, as ONE Inovalon, we are a united force delivering solutions that address healthcare’s greatest needs. Through our mission-based culture of inclusion and innovation, our organization brings value not just to our customers, but to the millions of patients and members they serve.

The Senior AI Development Engineer L4designs, builds, andoperatesfull-stack applications that integrate advanced AI/ML capabilities. The role focuses on agentic architecture, Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG) to deliver secure, scalable, enterprise solutions. The engineer partners closely with product, data science, and platform teams to deliver AI-powered experiences using Gemini Enterprise and Microsoft Copilot while adhering to responsible AI and regulatory requirements.

• Architect, develop, and maintain full-stack services and user interfaces integrating LLMs, agents, and RAG pipelines.
• Design agentic solutions (planning, tool-use, memory) and orchestration for multi-step workflows.
• Operationalize Gemini Enterprise and Microsoft Copilot integrations, including identity, permissions, and governance.
• Implement secure data grounding with vector databases (embeddings, chunking, indexing) and guardrails.
• Build evaluation frameworks for AI quality (precision/recall, hallucination checks, safety filters) and telemetry.
• Own CI/CD for AI-enabled services, including automated tests (unit/integration), canary deployments, and rollbacks.
• Collaborate with product teams and stakeholders to scope, estimate, and prioritize AI features and platform capabilities.
• Conduct performance tuning and cost optimization for LLM usage (token budgeting, caching, prompt design).
• Document architecture, APIs, prompts, and operational runbooks; train and support partner teams.
• Participate in design/code reviews, retrospectives, and continuous improvement initiatives.
• Ensure compliance with HIPAA, security, and Responsible AI policies (privacy, explainability, monitoring, incident response).
• Participate in on‑call rotation to support critical issues.
• Maintain compliance with company policies, procedures, and mission statement.
• Adhere to all confidentiality and HIPAA requirements as outlined in company operating policies.
• Fulfill additional responsibilities as reasonably assigned to support operational and financial success.
• Uphold responsibilities related to separation of duties for applicable processes and procedures.
• Understand that job description may change as business needs dictate; notice will be provided.

• 5+ years software development experience (Python, Type Script/JavaScript, C# or Java) with strong design/debug skills.
• Full‑stack frameworks (e.g., React/Next.js, Angular, .NET, Spring) and REST/Graph
  
  QL API development.
• Cloud deployment on AWS, Azure, and/or Google Cloud (Functions/App Service/Cloud Run, Kubernetes), IaC (Bicep/Terraform).
• Secure application development (auth
  
  N/auth
  
  Z, secrets, encryption, threat modeling) and compliance awareness.
• CI/CD (Git Hub Actions/Azure Dev Ops), automated testing, and monitoring (App Insights/Cloud Logging).

• Hands‑on experience with LLM platforms:
  Gemini Enterprise, Microsoft Copilot;
  OpenAI or Anthropic experience a plus.
• Agent frameworks and orchestration (e.g., Lang Chain/LCEL, Vertex AI Agents, Azure OpenAI orchestration, function/tool calling).
• RAG pipelines (document ingestion, text splitting, embeddings, retrieval strategies such as hybrid/BM25, re‑ranking).
• Vector databases (e.g., Big Query vector, Vertex Matching Engine, Azure Cognitive Search, Pinecone) and metadata schemas.
• Prompt engineering and safety guardrails (system prompts, tool descriptions, content filters,…