TuSimple is a self-driving technology company with a mission to bring automation to the trucking industry. Planning machine learning plays a critical role in the mission. At TuSimple, the planning ML team is responsible for developing the cutting edge machine learning algorithms to predict the future trajectories of the vehicles and pedestrians on the roads and model interaction between the ego truck with those dynamic environments. The planning ML model results are critical input to the downstream planning decision making modules to make optimized driving decisions.
Example projects we are working on include:
- Trajectory prediction using CNN technologies in environments that are hard to describe numerically
- Modeling the interaction between the truck and other road agents and predicting for the negotiation outcome
- Building solid benchmark system to quantify the data-driven model performance
- Encoding the prediction uncertainty and collaborate with planning to make stochastically optimized decision
What You'll Do:
- Deliver high-quality and reliable code for deep learning prediction modules on autonomous driving trucks
- Build and maintain efficient pipelines and reliable benchmarks for improving deep learning model performance
- Collaborate with other engineers to conduct system integration and tests.
What You'll Bring:
- 1-5+ years of professional experience working with autonomous vehicles.
- Masters in Computer Science, Math/statistics or other related fields.
- Product-quality code in C++ and/or Python.
- Experience in designing and building machine learning/deep learning data pipelines.
- Knowledge in deep learning topics including but not limited to CNN, RNN, Encoder-Decoder, GraphNN
- PhD in Computer Science, Math/statistics or related field.
- Knowledge and experience in neural network computation optimization, such as distributed training, active learning, network quantization, and CUDA acceleration.
- Experience with benchmarking and performance optimization of complex systems
- Experience with robotics simulation environments, model-based RL, and the use of model-predictive control to guide RL convergence
- Understanding of risk-aware planning concepts and algorithms, and general uncertainty management
- Familiar with front-end frameworks
- Visa sponsorship is available for this position
- Opportunity for professional growth and career advancement
- Competitive salary and benefits
- Daily breakfast, lunch, and dinner
- Shape the landscape of autonomous driving
- 100% Company paid Medical, Vision, and Dental insurance plan
- Company 401(K) program
- Company paid life insurance
- Company paid education/training.
- Company paid gym membership.
TuSimple is an Equal Opportunity Employer. This company does not discriminate in employment and personnel practices on the basis of race, sex, age, handicap, religion, national origin, or any other basis prohibited by applicable law. Hiring, transferring and promotion practices are performed without regard to the above-listed items.