Job Summary

We are looking for a Voice Systems Engineer to build and optimize real-time, GPU-accelerated voice AI systems using NVIDIA technologies. The role focuses on speech-to-text, text-to-speech, conversational AI, low-latency inference, and integrating NVIDIA speech/AI models into enterprise applications.

Key Responsibilities
- Build real-time voice AI pipelines for
speech-to-text, LLM reasoning, and text-to-speech
.
- Work with NVIDIA GPU technologies and optimize inference for low latency and high throughput.
- Integrate NVIDIA speech and generative AI models into enterprise AI platforms.
- Work with
NVIDIA NIM, NeMo, TensorRT, Triton Inference Server and CUDA/PyCUDA
.
- Build streaming voice APIs using Python, FastAPI, WebSockets/WebRTC or similar technologies.
- Optimize GPU utilization, concurrency,



batching and real-time inference performance.
- Integrate voice agents with enterprise applications, databases and APIs.
- Package and deploy production-ready voice AI solutions for enterprise environments.

Required Skills
- Robust
Python
and backend development experience.
- Experience with
speech AI / ASR / TTS / conversational AI
.
- NVIDIA GPU ecosystem experience:
CUDA, TensorRT, Triton, NIM or NeMo
.
- Experience with LLMs and real-time inference.
- WebSockets, WebRTC or streaming architectures.
- Docker and Kubernetes.
- Experience optimizing
latency, throughput and GPU utilization
.

📌 NVIDIA Voice Systems Engineer (Hyderabad)
🏢 nugget.ai
📍 Hyderabad