Application Guide

How to Apply for Staff Machine Learning Engineer

at Xpeng Motors

🏢 About Xpeng Motors

Xpeng Motors is a pioneering Chinese EV company that develops intelligent electric vehicles and eVTOLs (flying cars), with a strong focus on autonomous driving and AI. Working here means contributing to cutting-edge perception systems for next-gen mobility, in a fast-paced, innovation-driven environment.

About This Role

As a Staff Machine Learning Engineer on the perception team, you will own the development and optimization of 2D traffic sign detection models for autonomous driving. Your work directly impacts the safety and reliability of Xpeng's self-driving features, requiring deep expertise in computer vision, model deployment, and data-driven improvement.

💡 A Day in the Life

Your day might start with a stand-up to discuss recent detection failures from fleet data, then you'll dive into running training experiments with new data augmentations. After lunch, you could be debugging a model's false positives using visualization tools, and later collaborate with the calibration team to align camera parameters for better sign detection. The day ends with reviewing performance on a new validation set and planning next steps.

🚀 Application Tools

Generate Cover Letter

AI-powered, tailored for Xpeng Motors

Interview Prep

Practice questions for this role

🎯 Who Xpeng Motors Is Looking For

  • Has 3-5 years of hands-on experience with object detection models like YOLO, Faster R-CNN, or DETR, specifically applied to traffic signs or similar small, dense objects.
  • Demonstrates proficiency in PyTorch and experience with model optimization for production (ONNX, TensorRT, quantization) on embedded platforms.
  • Skilled in analyzing detection failures, curating training datasets, and improving metrics like mAP, precision/recall, and false positive/negative rates.
  • Holds a Master’s or PhD in Computer Vision, Robotics, or related field, with a track record of deploying models in real-world autonomous driving systems.

📝 Tips for Applying to Xpeng Motors

1

Tailor your resume to highlight specific traffic sign detection projects, including metrics improvements and deployment challenges solved.

2

Mention experience with TensorRT or ONNX runtime optimization in your cover letter; this is critical for production deployment in vehicles.

3

Include a link to your GitHub or portfolio showing open-source contributions or personal projects in detection or autonomous driving.

4

Research Xpeng's current ADAS features (e.g., XNGP) and mention how your work could enhance their traffic sign recognition.

5

If possible, get a referral from someone in the Xpeng AI or perception team via LinkedIn or industry conferences.

✉️ What to Emphasize in Your Cover Letter

['Emphasize your hands-on experience with detection model optimization for real-time, safety-critical systems (e.g., reducing latency while maintaining accuracy).', 'Highlight your ability to analyze failure cases and iteratively improve model performance using data curation and retraining.', "Show enthusiasm for autonomous driving and Xpeng's mission to create intelligent, eco-friendly mobility solutions.", 'Mention any cross-functional collaboration experience with software, hardware, or testing teams to deploy models in production.']

Generate Cover Letter →

🔍 Research Before Applying

To stand out, make sure you've researched:

  • → Read about Xpeng's XNGP (Navigation Guided Pilot) system and its current capabilities in traffic sign recognition.
  • → Look up recent papers or blog posts from Xpeng's AI team on perception challenges (e.g., CVPR, NeurIPS publications).
  • → Understand Xpeng's vehicle hardware (camera specs, compute platform like NVIDIA Orin) to tailor your optimization approach.
  • → Check Xpeng's news for any recent updates on their eVTOL project or autonomous driving milestones.

💬 Prepare for These Interview Topics

Based on this role, you may be asked about:

1 Walk me through how you would improve the detection of small or occluded traffic signs in a dense urban environment.
2 How do you handle dataset imbalance (e.g., rare sign types) and what augmentation strategies do you use?
3 Explain your experience with quantization and TensorRT optimization. How did it affect model accuracy and inference speed?
4 Describe a time you identified a systematic failure in a detection model and how you fixed it (e.g., false positives from similar objects).
5 How would you design an evaluation pipeline to catch regressions before deploying a new model to the vehicle fleet?
Practice Interview Questions →

⚠️ Common Mistakes to Avoid

  • Don't submit a generic resume without highlighting specific detection projects or metrics (e.g., mAP improvements).
  • Avoid claiming experience with deployment if you haven't actually optimized models for real-time inference on edge devices.
  • Don't ignore the importance of data curation and failure analysis; the job description emphasizes scenario analysis and dataset preparation.

📅 Application Timeline

This position is open until filled. However, we recommend applying as soon as possible as roles at mission-driven organizations tend to fill quickly.

Typical hiring timeline:

1

Application Review

1-2 weeks

2

Initial Screening

Phone call or written assessment

3

Interviews

1-2 rounds, usually virtual

✓

Offer

Congratulations!

Ready to Apply?

Good luck with your application to Xpeng Motors!

← Back to Job Listing Apply Now →