Algorithm Engineer, Deep Learning & Vision (New Grad)

Job description

### **Company Introduction** At Bot Auto, we are revolutionizing the transportation of goods with our cutting-edge autonomous trucks, enhancing the quality of life for communities around the globe. With the agility of a start-up and the wisdom of seasoned experts, Bot Auto boasts a team that has achieved numerous world-firsts and unparalleled innovations. United by a shared vision, we create miracles and propel the future of transportation. Join us and transform your dreams into reality. ### **Key Responsibilities** - **Model Implementation & Iteration:** Participate in the development, training, and optimization of state-of-the-art deep learning models for autonomous driving, with a focus on end-to-end architectures, including perception, online mapping, and end-to-end planning. - **Full Lifecycle Execution:** Engage in the entire machine learning workflow under the guidance of domain experts, spanning from data curation and data analysis to model experimentation, hyperparameter tuning, and rigorous performance metric verification. - **Cross-Functional Collaboration:** Partner with simulation, infrastructure, and downstream planning/control teams to deploy, evaluate, and integrate machine learning components into our production pipeline for autonomous trucks. - **Literature Tracking:** Stay abreast of the latest research breakthroughs in computer vision and generative AI, and actively bench-test promising SOTA methods to solve real-world corner cases. ### **How You'll Grow** This matters as much to us as what you'll ship. - **You get a real mentor.** Every engineer is paired with senior-level engineers developing you. Mentorship here is weighted toward design and judgment: how to frame a problem, what to build and why, how to tell whether a solution is actually right. - **We promote fast.** Managers are expected to push engineers to attempt work above their current level, and to promote in the next cycle when they deliver it. ### **Qualifications** ### Required**:** - **Education:** A Bachelor's, Master's, or Ph.D. (including upcoming graduates) in Computer Science, Robotics, Electrical Engineering, Applied Mathematics, Physics, or a related quantitative field. - **You have trained neural networks.** Coursework, research, personal projects, open-source work, and internships all count. We care that you have actually run the loop: built a model, trained it, found out why it was not working, and fixed it. - **Core Knowledge:** Strong theoretical foundation in machine learning and deep learning, with a solid understanding of modern architectures (e.g., Transformers, CNNs, Graphs). - **Technical Stack:** Proficiency in Python and deep learning frameworks such as PyTorch, along with strong software engineering fundamentals (data structures, algorithms, and clean coding practices). - **Attributes:** High self-motivation, strong analytical and problem-solving skills, a fast learner in a high-velocity startup environment, and a strong team-player mindset. ### Preferred**:** - **Computer vision.** Research or projects in computer vision, and particularly in 3D. - **Specific Research Directions:** Academic thesis or deeply focused research experience in one or more of the following domains: - Computer Vision (2D or 3D) - Online Mapping, Vectorization, or Visual SLAM - Prediction and Behavioral Modeling - **Academic Achievements:** A track record of research publications in machine learning, computer vision, or robotics conferences/journals (e.g., CVPR, ICCV, ECCV, NeurIPS, ICLR, ICRA, IROS). - **Engineering Plus:** Hands-on experience with model deployment, quantization, distillation, or inference acceleration tools (e.g., TensorRT, ONNX, CUDA, C++). - **Industry Exposure:** Prior internship experience within the autonomous driving industry or advanced robotics labs.