IND Staff Engineer, Reliability - The Hartford

Job Details

Location:

Hyderabad, Telangāna, IN

Category:

Information Technology

Employment Type:

Full time

Job Ref:

R2626209-333

We’re determined to make a difference and are proud to be an insurance company that goes well beyond coverages and policies. Working here means having every opportunity to achieve your goals – and to help others accomplish theirs, too. Join our team as we help shape the future.

Position Summary

We are seeking a highly skilled T7 AI Operations & Site Reliability Engineer to join our engineering team in Hyderabad, India. This role is laser-focused on the availability, reliability, and performance of our production AI systems. You will own the operational health of AI-powered products — ensuring LLM-based services, agentic workflows, RAG pipelines, and ML inference platforms maintain enterprise-grade uptime while scaling to meet demand. You will build the observability, automation, and incident response capabilities that keep our AI products running 24/7.

Level: T7 (Senior Engineer)

Location: Hyderabad, India

Employment Type: Full-Time

Key Responsibilities

AI Platform Reliability & Uptime

• Own the end-to-end reliability of production AI systems including LLM services, RAG pipelines, agentic workflows, and inference endpoints
• Define and maintain SLOs/SLIs/SLAs for AI products — latency, availability, error rates, token throughput, and response quality
• Design and implement high-availability architectures for AI workloads: multi-region failover, load balancing, auto-scaling, and graceful degradation
• Build circuit breakers, retry logic, fallback models, and rate-limiting strategies to ensure AI services remain available under stress
• Drive availability targets of 99.9%+ for critical AI-powered products
• Establish disaster recovery procedures and regularly test backup/restore for AI data stores and model artifacts

Observability & Monitoring

• Build and maintain comprehensive observability stacks for AI systems — metrics, logs, traces, and AI-specific signals (hallucination rates, model drift, token costs)
• Implement real-time dashboards and alerting for AI service health, model performance, and infrastructure utilization
• Design anomaly detection and proactive alerting to identify degradation before users are impacted
• Monitor LLM provider dependencies (GCP Vertex AI, OpenAI, etc.) and implement automated failover when external services degrade
• Track and optimize cost-per-inference, token utilization, and resource efficiency across AI workloads

Incident Management & Response

• Lead incident response for AI system outages and degradations — triage, mitigate, resolve, and communicate