Join Sarvam, a pioneering company building India's sovereign AI platform. We are developing a comprehensive AI stack encompassing research, models, infrastructure, and applications, dedicated to making AI truly impactful for India. Collaborating with leading enterprises and public institutions, Sarvam is backed by prominent investors like Lightspeed, Peak XV, and Khosla Ventures.

Our conversational AI platform, Samvaad, empowers enterprises, startups, digital natives, and governments. We are seeking experienced technical leaders to transform complex, undefined challenges into robust, production-ready systems. The platform currently manages millions of highly personalized voice interactions daily.

As the Staff Engineer for our core conversational AI system, you will be responsible for its entire lifecycle. This includes the data model, API design, system orchestration, and ensuring exceptional reliability and performance. You will champion high engineering standards and take full ownership of the backend for agent, campaign, and user management, utilizing Python and FastAPI.

You will implement workflow orchestration for intricate, long-running operations using Temporal. A key focus will be on performance and reliability for our multi-tenant SaaS platform, aiming for sub-second latency Service Level Objectives (SLOs). You'll also establish comprehensive observability through logging, metrics, and tracing to facilitate rapid debugging in production. Furthermore, you will develop robust integration testing infrastructure to enable fast, safe deployments that don't compromise customer-facing systems.

We are seeking an accomplished engineer with 8-12 years of experience in building production backend systems, possessing deep expertise in Python and FastAPI (or Golang). A mandatory requirement is hands-on experience in developing telephony infrastructure, specifically with SIP/VoIP technologies.

Your track record should demonstrate full-stack ownership, encompassing system design, trade-off analysis, and operational deployment. Proven experience with large-scale workflow orchestration, ideally with Temporal, is essential. You should have built and operated high-scale, multi-tenant, low-latency distributed systems that remain resilient under significant load.

Strong fundamentals in PostgreSQL and Redis, including schema design, query performance optimization, and caching strategies, are critical. We also value your judgment in making strategic decisions regarding build vs. buy, optimization vs. shipping, and abstraction vs. inlining. Experience with LLMs or AI/ML systems, along with prior involvement in early- or growth-stage startups, is considered a valuable asset.