Senior Software Engineer, Video Analytics

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA

Senior Software Engineer role focused on building large-scale distributed Vision AI platforms for video analytics using NVIDIA Metropolis. The role involves designing and developing functionalities for video processing, integrating VLMs, CV models, and LLMs, and optimizing performance on NVIDIA hardware. Requires strong software development experience with ML systems, C++, Python, and GPU acceleration.

What you'd actually do

  1. Lead the creation of modern software, services, systems, and AI agents for video processing across diverse domains, including smart cities, indoor spaces, and industrial environments.
  2. Engage in the complete software lifecycle—from conceptualization and prototyping to development, accuracy & performance tuning, and production release.
  3. Explore, optimize, and integrate innovative technologies, including Vision-Language Models (VLMs), Computer Vision (CV) models, and Large Language Models (LLMs), to implement forward-looking video processing capabilities on NVIDIA hardware.
  4. Apply a strong software background to incorporate agility and rigor into design using AI-assisted coding capabilities, ensuring the highest degree of reliability and maintainability.
  5. Evaluate and fine-tune models using advanced tooling to improve accuracy and hardware utilization.

Skills

Required

  • modern C++
  • Python
  • Linux
  • algorithms
  • data structures
  • concurrency
  • distributed systems concepts
  • multimodal VLMs
  • LLMs
  • machine learning inference
  • GPU acceleration
  • CUDA
  • TensorRT
  • PyTorch
  • microservices
  • distributed architectures
  • REST APIs

Nice to have

  • video AI solutions
  • multimedia technologies
  • codecs
  • streaming pipelines
  • NVIDIA TAO
  • AutoML
  • multimodal datasets
  • embedded environments
  • NVIDIA Jetson
  • Orin platforms
  • Docker
  • Kubernetes
  • Helm
  • Azure
  • AWS

What the JD emphasized

  • production-grade machine learning systems
  • multimodal VLMs and LLMs
  • machine learning inference using GPU acceleration
  • video AI solutions

Other signals

  • large-scale distributed Vision AI platforms
  • video analytics services and solutions
  • integrate innovative technologies, including Vision-Language Models (VLMs), Computer Vision (CV) models, and Large Language Models (LLMs)