Cognizant

Senior AI Engineer – Claims Summarization & Agentic AI

Cognizant
Apply
5 days ago
Toronto, CanadaSenior

Base Salary

$110k - $130k/yr

Responsibilities

  • Design and build claims summarization agents for structured and unstructured claims data.
  • Develop multi-agent workflows for claim intake, history review, adjuster-note analysis, fraud indicators, and settlement recommendations.
  • Build agent orchestration using LangGraph, AutoGen, LangChain, CrewAI, or Databricks Mosaic AI Agent Framework.
  • Integrate and optimize enterprise LLMs, including Azure OpenAI, Anthropic Claude, Google Gemini, and open-source models.
  • Perform prompt engineering, evaluation, guardrail implementation, model tuning, confidence scoring, and hallucination mitigation.
  • Develop RAG solutions with Databricks Vector Search, Azure AI Search, embeddings, and knowledge graphs.
  • Implement document chunking, indexing, metadata enrichment, and retrieval optimization for claims documents.
  • Build scalable Azure Databricks and PySpark AI pipelines using Databricks Mosaic AI, MLflow, Unity Catalog, Delta Lake, and LakeFlow Pipelines.
  • Manage AI deployment, monitoring, governance, observability, privacy, explainability, auditability, and regulatory compliance.
  • Collaborate with claims subject-matter experts, product owners, data architects, engineering teams, and stakeholders while mentoring junior engineers.

Requirements

  • 8+ years of software engineering, data engineering, or AI engineering experience.
  • 3+ years of hands-on experience with generative AI and LLM solutions.
  • Proven experience implementing enterprise-scale AI agents and RAG-based applications.
  • Hands-on experience with LangGraph, LangChain, AutoGen, CrewAI, LlamaIndex, and Databricks Mosaic AI Agent Framework.
  • Experience with Azure OpenAI, Anthropic Claude, Google Gemini, open-source LLMs such as Llama and Mistral, Azure Databricks, Azure AI Search, Azure Storage or ADLS, Azure Functions, Azure DevOps, Delta Lake, Unity Catalog, and MLflow.
  • Proficiency in Python, PySpark, and SQL.
  • Experience with prompt engineering, RAG, vector databases, embeddings, agent orchestration, model evaluation, fine-tuning, and responsible AI.
  • Insurance domain experience, particularly in claims, policy administration, or underwriting, is highly preferred.
  • Preferred certifications include Microsoft Certified: Azure AI Engineer Associate, Microsoft Certified: Azure AI Apps & Agents Developer Associate, Databricks Certified Machine Learning Professional, Databricks Certified Data Engineer Professional, and Azure Data Engineer Associate.
  • Strong stakeholder communication, presentation, problem-solving, architecture, collaboration, and technical leadership skills.

Benefits

  • Hybrid work arrangement in Toronto.
  • Extended Group Health Plan through Manulife, prescription drug coverage, virtual healthcare through TELUS Health Virtual Care, and TELUS Virtual Pharmacy.
  • Mental health and counseling services, Employee Assistance Program, Mental Health Ally Program, wellbeing resources, events, and champion networks.
  • Life Insurance, Accidental Death & Dismemberment Insurance, and leave-of-absence and disability support programs.
  • Group Registered Pension Plan for associates on Canada payroll, employer matching contributions subject to plan terms, Group RRSP, and competitive group investment rates through Manulife.
  • Application accepted through 9/31/2026.

Tech Stack

AzureDatabricksMLflowPythonSQL

Categories

Cognizant

About Cognizant

10,000+ employees

Cognizant is a public IT services and consulting firm that designs, builds, and runs enterprise technology, including digital engineering, cloud modernization, data/AI, and managed services. It sells consulting, systems integration, and outsourcing on multi-year engagements to large enterprises in healthcare, banking, retail, communications, and manufacturing. Founded in 1994 and headquartered in Teaneck, New Jersey, Cognizant is NASDAQ-listed (CTSH) and a Fortune 500 company.

Contact me