This role pays more than
71%
of similar roles. Most pay
$165,412–$246,706
— the shaded band above.
At the midpoint, this role pays about
$236k
versus about
$206k
for comparable roles.
Based on 240 similar postings.
Employer
About Nvidia
Nvidia is a leading designer of graphics processing units (GPUs) and system-on-chip units, powering gaming, professional visualization, data centers, and artificial intelligence workloads. Industry: Semiconductors & AI Computing
Nvidia currently has
892 open roles
on FindRole.
Listed pay typically runs
$184,000–$287,500
across 870 roles with salary data.
Senior Context Fusion AI Engineer - Autonomous Vehicles joins the DRIVE Road Structure, Online Mapping, and Context Fusion team to develop a learned 3D/4D world model for L3/L4 autonomous driving. The role involves building multimodal sensor-fusion systems that integrate camera, LiDAR, radar, and vehicle state data into a unified spatiotemporal representation. Key responsibilities include developing architectures for road graph elements, occupancy perception, and motion prediction while creating training pipelines for large-scale datasets. Candidates will utilize PyTorch, C++, Python, CUDA, and NVIDIA hardware to implement Transformer-based models, vision-language models, and BEV representations. The position focuses on solving the technical challenge of creating a robust, uncertainty-aware representation that supports road understanding and planning in complex environments. Essential skills include 3D geometry, coordinate transforms, and experience with distributed training and inference optimization for production-quality autonomous vehicle systems.
Develop learning-based multimodal sensor-fusion systems to create a unified spatiotemporal world representation from camera, LiDAR, radar, and vehicle state inputs.
Build architectures that jointly reason over multi-sensor data while handling calibration, synchronization, coordinate transforms, and uncertainty.
Create models to output road graph elements, semantic scene understanding, and occupancy representations for autonomous driving.
Design scalable multimodal fusion architectures using Transformer-based methods, BEV representations, and temporal context aggregation.
Build training, fine-tuning, and evaluation pipelines for large-scale multimodal datasets.
Investigate foundation-model approaches including vision-language models and multimodal pre-training for autonomous navigation.
Develop systematic analysis and debugging tools to identify model failures and cross-sensor disagreements in simulation and on-road testing.
Translate research advances into production-quality systems by defining multi-task objectives that balance perception, geometry, and safety.
What we're looking for
BS, MS, or PhD in Computer Science, Robotics, Electrical Engineering, Machine Learning, or a related technical field, or equivalent experience.
8+ years of experience, including at least 2+ years in the AV or robotics industry and 2+ years of leadership experience in a technical area.
Strong experience developing production-quality sensor-fusion, perception, state-estimation, or autonomous-driving systems.
Demonstrated experience with learning-based multimodal perception or fusion involving camera, LiDAR, radar, map, navigation, and ego-motion signals.
Solid understanding of 3D geometry, coordinate frames, calibration, temporal synchronization, ego-motion compensation, tracking, uncertainty estimation, and sensor failure modes.
Experience with deep-learning methods for 3D perception, scene representation, occupancy prediction, semantic segmentation, object detection/tracking, motion prediction, or planning.
Strong C++ and Python programming skills with hands-on experience training and optimizing models in PyTorch.
Experience with Transformer, VLM, or multimodal foundation-model architectures (preferred); experience with BEV, point-cloud/voxel, or neural scene representation (preferred).
C++
Python
CUDA
TensorRT
PyTorch
Linux
QNX
Machine Learning
Model Inference
Quantization
Real-time Operating Systems
Autonomous Driving
ADAS
Robotics
performance-analysis
Software Architecture
PyTorch
Python
C++
CUDA
Deep Learning
Computer Vision
Transformers
CNN
3D Perception
BEV
Multi-sensor Fusion
LoRA
Vision-Language Models
Active Learning
Autonomous Driving
Reinforcement Learning
Python
C++
Deep Learning
GPU Clusters
Kubernetes
SLURM
High-Performance Computing
Data Pipelines
Simulation
Closed-Loop Simulation
PPO
GRPO
NuRec
Cosmos World Model