Perception Intern

Job Description

Perception Intern, Summer 2026 | Mach9

The Tone:
This is an internship at Mach9. Mach9 builds products that leverage advanced 2D image and 3D point cloud models to enhance the development and management of urban infrastructure. This role is critical as you will contribute to core product development, helping to shape the future of urban infrastructure management through impactful projects at the intersection of machine learning and geospatial technology.

The TL;DR
• Role: Internship
• Type: Temporary
• Location: In-person, San Francisco, CA

• Mission: Contribute to the development and enhancement of Mach9’s core products by applying perception models to urban and roadway features.
• Tech Stack: Python, PyTorch, Jax

What You’ll Actually Do
• Contribute: Engage in a range of projects that are critical to the development and enhancement of Mach9’s core products.
• Analyze: Work on advanced 2D image and 3D point cloud models for object and panoptic detection, specifically focusing on urban and roadway features like traffic lights, utility poles, curbs, and painted road markings.
• Prototype: Develop initial prototypes, such as those to detect paint lines and curbs on roadways, which can evolve into new product offerings.
• Optimize: Implement and adapt perception models and algorithms for real-time detection, balancing accuracy and speed to run efficiently on advanced edge mapping hardware systems.
• Tune: Assist in the development and fine-tuning of machine learning models to enhance their overall performance and accuracy.

The Must-Haves
• Background: Student currently pursuing a degree in Machine Learning, Computer Science, or a related field, possessing a solid understanding of machine learning concepts and techniques.
• Experience: Hands-on experience with machine learning frameworks such as PyTorch or Jax, and familiarity with computer vision tasks.
• Skills: Demonstrate strong programming fundamentals in Python, coupled with a focus on algorithmic problem-solving and clean code design; possess the ability to quickly implement and iterate on heuristic or geometry-based algorithms; exhibit excellent problem-solving abilities and the capacity to work independently on complex tasks.
• Bonus: A keen interest in applying machine learning to solve real-world problems, particularly in the context of geospatial data analysis, and a collaborative mindset.