Product Manager, Model APIs

Sarvam is building the bedrock of Sovereign AI for India. The company is developing India's full-stack sovereign AI platform, building across research, models, infrastructure and applications with a singular focus on making AI genuinely work for India. Sarvam works with leading enterprises and public institutions and is backed by Lightspeed, Peak XV, and Khosla Ventures. Sarvam partners with India's leading brands, including Tata Capital, SBI Life, CRED, IDFC, and LIC.

We are hiring a Product Manager for Model APIs. This is the person who owns how developers and enterprises consume Sarvam's models, and who carries what they learn back into what we train next.

You will sit at the seam between research and the market. On one side, you own the API surface: what we ship, how it is shaped, how it is priced and packaged, and whether it holds up in production. On the other, you help set model direction: which capabilities to prioritize, which languages and domains to invest in, what “good enough to ship” means, and where the next release should aim. This is a role for someone who wants to influence the model roadmap itself, not just wrap it in an interface.

• Own the product surface for Sarvam's model APIs across speech, language, and multimodal, from spec and versioning to documentation, developer experience, and deprecation.

• Shape model roadmaps alongside research leads, bringing evidence from usage, evals, and customer signal into decisions about what gets trained, fine-tuned, and released.

• Define what quality means for each model line and hold the bar: accuracy, latency, cost per call, robustness, and the failure modes that matter to real deployments.

• Own evaluation as a product, including benchmark suites, offline and online tests, human evals, and the reporting that makes model progress legible to everyone from researchers to customers.

• Turn customer and developer feedback into concrete model and API requirements, and close the loop by telling them what changed and why.

• Partner with infra and platform teams on rollout, capacity, monitoring, and reliability for models in production.

• Drive release planning end to end: positioning, launch, migration paths for existing users, and the trade-off calls that come with each.

• Communicate priorities and trade-offs clearly across research, engineering, GTM, and leadership.

• 3+ years building AI/ML products, developer platforms, or APIs, ideally where the underlying model was part of the product decision rather than a fixed input.

• Working fluency with the model lifecycle: data, training, fine-tuning, evaluation, deployment, and monitoring. You do not need to train models, but you should be able to hold your own in a conversation with the people who do.

• Strong analytical instincts and genuine comfort with data. You reach for the numbers before the narrative.

• Sound judgment on trade-offs between quality, latency, and cost, and the confidence to make a call and defend it.

• Ability to work closely with researchers and earn their trust.

• Experience with speech (ASR, TTS) or LLM APIs, and the metrics that govern them: WER, CER, MOS, TTFT, tokens per second.

• Track record shipping and maintaining a public or enterprise-facing API.

• Hands-on with SQL, Python, or ML tooling, enough to run your own analysis.

• Exposure to Indian language AI, multilingual systems, or evaluation for low-resource languages.

Sarvam is a fast-moving, high talent-density team building full-stack AI for India, working on problems that push the frontiers of AI with real population-scale impact.

• Work alongside researchers, engineers, builders, and business leaders who move fast and hold each other to a very high bar

• High ownership and high impact, from day one

• Everything we do is AI-first, from the way we build and ship to the way we think about problems

• You can work on problems that could change how an entire country learns, works, and communicates

If you want to work on problems at the frontier of AI in India, Sarvam is the place to be.