Description
Sarvam is hiring a Backend Engineer to own the architecture and production operation of its OCR and document-extraction serving harness. The role covers API and orchestration layers, multi-pass inference, model routing, verification, batching, GPU utilization, Temporal workflows, autoscaling, observability, multi-tenancy, and cost optimization for national-scale document processing. Candidates should have 5–6+ years of backend engineering experience, strong Go or Python skills, distributed systems expertise, production experience with Temporal or similar durable execution engines, and experience serving ML or LLM inference.
