Frontend Engineer, Vision

About the role

Join Sarvam’s Vision team to build the user interfaces that let humans see exactly what AI models see in complex Indian documents. You will create document viewers, correction workflows, schema designers, and evaluation dashboards that are performance‑sensitive and stateful.

Key responsibilities

• Build document review UI: PDF/image rendering, page navigation, bounding‑box overlays, confidence highlighting, and side‑by‑side source‑to‑extraction linking.
• Create human‑in‑the‑loop correction workflows with fast keyboard‑driven review, field‑level edit, approve/reject queues, and feed corrections back into the evaluation loop.
• Develop an extraction schema designer allowing users to define, test, and version structured output for document types.
• Implement internal evaluation and observability dashboards showing accuracy by field, latency, cost breakdowns, and model comparisons.
• Handle real‑time and long‑running state: streaming partial results, job progress, optimistic updates, and resumable sessions.
• Own frontend performance for heavy documents using virtualization, canvas rendering, lazy loading, and memory discipline.

Required profile

• 4–5 years of experience building production frontend applications with strong ownership.
• Deep expertise in React (hooks, rendering behavior, state management at scale) and performance profiling.
• Strong TypeScript skills, using the type system to prevent bad states.
• Solid understanding of the browser platform, layout, events, memory, and custom rendering with Canvas or SVG.
• Experience with complex, data‑dense product surfaces such as dashboards, editors, or annotation tools.

Required skills

• React
• TypeScript
• JavaScript
• Canvas / SVG rendering
• PDF rendering in the browser (e.g., PDF.js)
• WebSockets or Server‑Sent Events for streaming
• Python
• Go
• Kubernetes
• PostgreSQL
• Redis
• OpenTelemetry
• Temporal
• REST APIs