Senior Audio Research Engineer
Bangalore · Onsite
About the Role
Building India’s most powerful custom SLMs and real-time Speech-to-Speech voice agents that already handle millions of calls across banking, healthcare, and new economy brands.
We want to take our platform at even crazier scale. We are building things that real people use. Come deploy models you’re proud of.
What You'll Do
- Architect infrastructure for custom SLMs and real-time Speech-to-Speech voice agents
- Build and scale custom diffusion-based audio models (cutting-edge, not just huggingface copy-paste)
- Optimize inference until it can handle 10,000+ requests per second
- Deploy models using TorchScript, ONNX, TensorRT, and DeepSpeed
- Wrestle with GPUs and squeeze every millisecond out of the pipeline
What We're Looking For
- Experience building or contributing to diffusion-based audio models
- Deep expertise in PyTorch (it should be your second language)
- War stories deploying models to production (TorchScript/ONNX/TensorRT)
- Experience optimizing inference for massive scale (10,000+ RPS)
- Ability to squeeze every millisecond out of a pipeline
Benefits
- ₹70 lakh cash + significant ESOPs (skin in the game)
- Work onsite in Bengaluru
- Ownership over models that touch real people, not just “demo” users
- Work with a smart, hungry, no-BS team
Interested?
Please share your resume on LinkedIn or email at shubham@mowka.in