Backend Engineer - Studio Media Platform
Bengaluru • FullTime
Posted 3mo ago
About the job
Sarvam is building the bedrock of Sovereign AI for India, developing a full-stack AI platform across research, models, infrastructure, and applications. We partner with leading enterprises and public institutions, backed by prominent venture capital firms. We are hiring a Backend Engineer to work on our Studio media platform, which includes AI dubbing, live translation, and foundational services for voice cloning, stem separation, lip sync, and music generation. You will build and maintain production services, ML pipeline libraries, and platform SDKs to enable multilingual media processing at scale for our enterprise customers and Studio users.
Responsibilities
- Design and optimize production FastAPI services for dubbing and live translation, including task orchestration, scheduling, and backpressure controls.
- Build and maintain distributed worker architectures with independent scaling and automatic recovery for tasks.
- Own and manage the data layer, including async ORM models, schema migrations, and query optimization on PostgreSQL.
- Implement real-time features such as WebSocket-based job tracking and streaming audio pipelines.
- Manage Kubernetes deployments using Helm charts, secrets management, and ingress configuration.
- Extend and maintain the core dubbing library across all pipeline stages.
- Integrate and optimize ML model serving, including remote and local inference.
- Build and improve QC orchestration for automated scoring, tempo analysis, and pronunciation verification.
- Design async-first pipelines with efficient concurrency patterns for audio processing.
- Maintain and evolve LLM integration layers for translation, QC, and pre-processing.
- Build and maintain the shared Studio service SDK with reusable middleware and routers.
- Design media storage abstractions for upload, signed URL generation, and cloud blob storage integration.
- Implement cross-cutting concerns like rate limiting, metering, audit trails, and request history.
- Build observability foundations with OpenTelemetry instrumentation, structured logging, and metrics collection.
- Maintain high test coverage with strong CI gates and thorough mocking of external services.
- Instrument services with custom metrics and structured error tracking for production observability.
- Manage CI/CD pipelines for automated testing, container builds, and version management.
Requirements
- 4-6 years of experience in backend engineering, focusing on building and operating production services at scale.
- Strong proficiency in Python with hands-on experience building production FastAPI or similar async web services.
- Deep understanding of async programming, including asyncio, concurrent execution patterns, and high-throughput workloads.
- Experience with distributed task systems, task queues, message brokers, and fault-tolerant job orchestration.
- Hands-on experience with PostgreSQL and an async ORM (SQLAlchemy preferred), including query optimization, schema design, and migrations.
- Familiarity with audio/media processing tools and libraries like FFmpeg, soundfile, or librosa.
- Experience integrating ML models into production via APIs or local inference.
- Experience building reusable libraries or SDKs, including API design and backward compatibility.
- Proficiency with Docker, Kubernetes, Helm charts, and at least one major cloud platform (Azure/GCP/AWS).
- Strong testing discipline, including writing thorough tests, mocking external services, and maintaining CI/CD pipelines.