About this role
Gatik needs a machine learning engineer to build, optimize, and deploy production models that power autonomous middle-mile delivery trucks. The incoming engineer will own the entire machine learning lifecycle from data strategy and model training to real-time deployment on vehicle hardware and cloud infrastructure. Daily work involves designing efficient neural networks for perception, prediction, and planning while collaborating closely with perception, systems, and hardware teams to meet strict latency constraints.
Ideal candidates are pragmatic engineers who enjoy bridging the gap between offline research and high-performance C++ autonomous systems. Someone with a strong background in model optimization and a track record of deploying robust machine learning models into performance-constrained environments will thrive in this collaborative setting.
What you will do
- Manage the complete machine learning lifecycle including data preprocessing, training, evaluation, optimization, deployment, and monitoring.
- Develop and refine neural network models that support perception, prediction, planning, and scene understanding for autonomous vehicles.
- Optimize models using quantization, pruning, compression, and efficient architecture design to meet strict latency and compute constraints.
- Integrate trained models into C++-based autonomy systems and optimize inference for production vehicle hardware using CUDA and TensorRT.
- Analyze model performance using simulation and real-world driving data to identify failure modes and drive continuous improvements.
- Build high-throughput pipelines for training, evaluation, data processing, and large-scale offline inference.
- Partner with autonomy, systems, hardware, and infrastructure teams to ensure machine learning components integrate reliably into the vehicle platform.
What you need
- Hold an MS or PhD in Computer Science, Machine Learning, Robotics, Electrical Engineering, Statistics, Optimization, or a related field.
- Demonstrate strong Python programming skills along with experience using frameworks like PyTorch or TensorFlow.
- Bring strong C++ skills with proven experience integrating machine learning models into high-performance production systems.
- Possess a deep understanding of machine learning workflows including data curation, training, evaluation, and inference optimization.
- Show experience deploying and optimizing neural networks for real-time, embedded, robotics, or autonomous driving systems.
- Demonstrate experience with model optimization techniques such as quantization, pruning, compression, and efficient architectures.
- Bring experience with software architecture, profiling, latency optimization, system-level debugging, and data flow analysis.
Nice to have
- Experience with transformers, multimodal models, diffusion models, world models, or end-to-end driving models.
- Experience in autonomous driving, robotics, or other safety-critical real-time machine learning systems.
- Publications or demonstrated technical contributions in efficient machine learning, autonomous robotics, or related areas.
- Prior contributions to large-scale machine learning systems deployed in production.
Skills and tech stack
Questions candidates ask
What is the work location and remote policy for this role?
This role is onsite 5 days a week at the Santa Clara, CA office.
What is the salary range for this machine learning position?
The salary range for this role is $170,000 - $240,000.
How is the experience level determined for applicants?
The company is open to all experience levels, and leveling will be determined based on experience and technical depth.
What kind of systems and tech stack will I be working with?
You will work with Python, C++, PyTorch, TensorFlow, CUDA, TensorRT, and Azure to deploy models into C++-based autonomy systems.