SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Sarvam AI is building India's full-stack sovereign AI platform, backed by Lightspeed, Peak XV, and Khosla Ventures. The company develops foundation models and infrastructure serving leading Indian enterprises including Tata Capital, SBI Life, CRED, IDFC, and LIC.
You will own major end-to-end areas of Sarvam's model API platform, which handles tens of millions of API calls daily across ASR, TTS, Vision, LLM, and Translation services. Working closely with the Staff Engineer leading platform architecture, you will take ownership of critical areas such as speech-to-text pipelines, text-to-speech pipelines, or WebSocket streaming layers, and be accountable for their reliability, performance, and end-to-end latencies.
Key responsibilities include:
- Own one or more major platform components end-to-end: build, operate, and continuously improve them
- Convert loosely defined goals into RFCs, milestones, and shipped results; document decisions and trade-offs
- Be on-call post-launch for incident response, root-cause analysis, and follow-through on fixes
- Make critical decisions and own outcomes
- Obsess over latencies: dissect existing services, understand system behavior, and tune performance knobs
- Bring services online without regressions in customer-facing APIs, latency, or reliability
- Build and operate streaming infrastructure: WebSocket connections, long-lived sessions, queue-based batch processing, backpressure, and low-latency model invocation
- Engineer audio pipelines: TTS chunking around model context limits, sample-rate conversion, format encoding, and VAD-based silence detection for ASR
- Hold services to sub-100ms latency SLOs in a multi-tenant environment
- Reason from first principles and constraints: latency budgets per hop, throughput, cost per request, failure modes, and tenant isolation
- Use back-of-the-envelope estimates to decide what to build, optimize, or leave alone
- Build in observability from day one: logging, metrics, and tracing for rapid production debugging
- Strengthen integration testing harness to enable fast shipping without breaking customer-facing APIs
- Partner with Inference and MLOps teams on fast, reliable, operable model serving
- Mentor junior engineers through RFC reviews, code review, and pairing
- Practice design-first, documentation-first engineering: RFCs before code, no tribal knowledge
Requirements:
- 4–6 years of building and operating production backend systems
- At least one service owned end-to-end under real load, including understanding what broke and what you changed
- Strong distributed-systems fundamentals: concurrency, networking, queues, caching, consistency, and failure modes
- Experience with low-latency or high-throughput services; comfort reasoning about p95 and p99 latency
- Solid fundamentals with relational databases and caches (e.g., PostgreSQL, Redis): schema design, query performance, caching strategy
- Fluency in at least one backend language with willingness to work in both Python and Golang
- Experience running services with containers, CI/CD, and production monitoring
- Daily, hands-on use of AI tools in engineering workflow with concrete examples of automation or speedup
- Bias for action: fast on reversible decisions, careful on irreversible ones
- Clear writing, especially RFCs and design docs that others can act on
Good to have (not required):
- Voice AI or real-time audio experience: WebSockets, WebRTC, streaming ASR/TTS, or audio processing with FFmpeg/pyav, codecs, sample rates, VAD
- Production experience in Go or Python with FastAPI
- Experience serving LLM, ASR, TTS, or vision models in production
- Experience building metering, billing, wallet, or payments infrastructure
- Running services on Kubernetes in production
- Time at an early- or growth-stage startup
