Perception Intern (Fall 2026)

Zipline•South San Francisco, CA

3d•$50•Onsite

About The Position

Zipline is the world’s largest and most experienced drone delivery service, aiming to serve all humans equally by ensuring access to food, medicine, and essential goods. They design, build, and operate the world’s largest autonomous logistics system, making deliveries globally. The company emphasizes practical problem-solving, real-world challenges, and rapid growth, with a team motivated by building systems that have a direct, meaningful impact. The Droid Perception Team specifically focuses on onboard and offboard perception systems that inform, validate, and augment the aircraft's autonomy. This involves preparing Zipline aircraft for mission-critical deliveries in complex environments through work on 3D and semantic priors, customer preferences, and terrain features. The intern will dive into ML model experimentation, evaluation, and integration, pushing the boundaries of the offboard perception system. This role offers the opportunity to see ideas transition from concept to real-world application in a fast-paced, collaborative environment. The intern will research current field advancements to propose novel solutions to perception challenges, contributing to the future of autonomous deliveries.

Requirements

Must have completed the second year of undergraduate studies. Masters and PhD students are also eligible.
A proven history of quality ML research, reflected by publications in conferences such as NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, CoRL, RSS, etc.
Specific expertise in areas such as multi view depth estimation, semantic segmentation, generative modeling, etc.
Proficiency with deep learning frameworks like PyTorch, JAX, or TensorFlow.
Experience writing code in a production environment, e.g. at a previous internship.
Ability to rapidly experiment, iterate, and adapt to new findings in a dynamic environment.

Responsibilities

Ideate, experiment and iterate on learning-based solutions for unique perception challenges.
Collaborate closely with team members, brainstorming and deriving creative solutions from first principles.
Leverage heterogeneous sources of data (real-world, simulation and internet-scale data) to train machine learning models.
Ship production code to train, validate and integrate models into the perception system.
Share findings and insights, fostering knowledge exchange across teams.