AI Architect– Insurance (Mandatory) | Azure | API-First Microservices (.NET at TMS LLC

Jersey City, New Jersey, United States -

Full Time

Start Date

Immediate

Expiry Date

14 Jun, 26

Salary

0.0

Posted On

16 Mar, 26

Experience

10 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

AI Architecture, Azure, API-First, Microservices, .NET, MLOps, LLMs, GenAI, NLP, Underwriting, Pricing, Claims, Fraud Detection, Responsible AI, Docker, Kubernetes

Industry

IT Services and IT Consulting

Description

Company Description Job Description Role: AI Architect– Insurance (Mandatory) | Azure | API-First Microservices (.NET Program) Duration: Long Term Location: Remote/ EST Experience: 15+ years overall; 4+ years in AI/ML architecture/engineering Role Summary We are building a next-generation insurance platform, including a greenfield P&C Policy Administration System (PAS) with a microservices-based, API-first architecture on Microsoft .NET. As the AI / ML Architect, you will lead the design and delivery of AI-powered capabilities across underwriting, pricing, claims, fraud, and operations. You will define end-to-end AI architecture (data → model → MLOps → serving), ensure secure and compliant AI, and partner closely with product, actuarial, underwriting SMEs, and engineering teams to move from prototypes to production-scale AI. Insurance domain experience is mandatory for this role. Key Responsibilities 1) AI Architecture & Solution Design (End-to-End) Define the target-state AI/ML architecture for insurance use cases: underwriting decision support, risk scoring, claims triage, fraud detection, pricing optimization, customer/agent assist, and personalization. Select and guide model approaches: predictive ML, LLMs/GenAI, NLP (and vision models where applicable), with clear tradeoffs and success metrics. Design API-first AI services that integrate cleanly with microservices (REST/gRPC, event-driven triggers, idempotency, versioning). Define patterns for feature pipelines, model serving, and governance that work across multiple pods and environments. 2) Model Engineering, MLOps & Deployment (Production Focus) Lead model development lifecycle: training, evaluation, validation, release, monitoring, and periodic refresh. Implement MLOps pipelines: automated model testing, monitoring, drift detection, model registries, approval workflows, and rollback strategies. Define serving patterns (batch/real-time/streaming) and optimize for accuracy, latency, reliability, and cost. 3) Insurance Domain Alignment (Business + Actuarial + Underwriting) Partner with product owners and translate requirements into AI-enabled components and measurable outcomes. Ensure AI outputs comply with underwriting guidelines, rating practices, claims workflows, and internal governance. Design human-in-the-loop controls where needed for regulated decisioning and operational safety. 4) Responsible AI, Security, Compliance & Risk Establish responsible AI guardrails: explainability, fairness/bias mitigation, audit trails, traceability, and model documentation standards. Ensure data privacy/security controls across the pipeline: PII handling, access controls, encryption, secrets management, and environment separation. Collaborate with risk/compliance to meet insurance regulatory expectations for AI systems (governance, reproducibility, reviewability). 5) Platform Integration & Cross-Functional Leadership Work closely with the Chief Architect, .NET architects, data architect, DevOps, and engineering pods to align AI services to platform standards. Mentor data scientists/ML engineers; enforce engineering rigor (testing, reliability, monitoring, secure coding). Drive POCs and technology evaluations, and productize successful capabilities into reusable platform services. 6) AI-Assisted Engineering Enablement (Claude Code, Cursor, MCP) Use Claude Code and Cursor as first-class development accelerators (code generation, refactoring, test generation, documentation), with strong review and security guardrails. Standardize patterns for tool usage across teams, including MCP-based workflows/integrations (where applicable), ensuring traceability and quality gates. Define measurement for productivity and quality improvements (cycle time, rework, defect leakage, release stability). Must-Have Qualifications Insurance Domain (Mandatory) Proven insurance industry experience is required (P&C preferred): underwriting, rating/pricing, claims triage, fraud, policy servicing, or insurance data/analytics. Experience designing or integrating ML/AI solutions in insurance decisioning contexts (e.g., risk scoring, pricing, fraud, claims). Technical (Azure-first) 4+ years hands-on AI/ML engineering and/or architecture experience; overall experience typically 12+ years. Strong experience with Azure AI ecosystem, including one or more of: Azure Machine Learning (training, registries, endpoints) Azure OpenAI / LLM integration patterns Azure AI Services (language, vision, etc.) Strong MLOps experience: CI/CD for ML, model registries, monitoring, drift detection, evaluation, and controlled rollouts. Experience building API-first services and deploying ML systems using Docker and Kubernetes (AKS preferred). Engineering & Collaboration Strong communication skills: can explain model tradeoffs and risks to non-technical stakeholders and client executives. Proven ability to lead cross-functional teams in fast-paced environments and ship production outcomes. Strong P&C insurance experience (Auto/Home/Commercial) and familiarity with PAS workflows. Experience with event streaming (Kafka/Event Hubs) and real-time inference/feature pipelines. Experience with responsible AI frameworks and interpretable ML methods in regulated environments. Azure certifications (Azure AI Engineer / Azure Solutions Architect). Additional Information All your information will be kept confidential according to EEO guidelines.

Responsibilities

The AI Architect will lead the design and delivery of AI-powered capabilities across insurance functions like underwriting, pricing, and claims, defining the end-to-end AI architecture from data to serving.