Machine Learning Engineer Intern, Perception at Zoox

Boston, Massachusetts, United States

Apply Now

Not SpecifiedCompensation

InternshipExperience Level

InternshipJob Type

UnknownVisa

Autonomous Vehicles, RoboticsIndustries

Requirements

Currently working towards a B.S., M.S., Ph.D., or advanced degree in a relevant engineering program
Must be returning to school to continue your education upon completion of the internship
Demonstrate outstanding academic performance
Activities outside of coursework
Aptitude
Curiosity
Passion for Zoox's mission

Responsibilities

Develop advanced multimodal large language models to enhance scenario understanding and driving (Offline Driving Intelligence team)
Develop and fine-tune models with driving data to identify hazards, interpret driving restrictions, and answer questions about scenarios (Offline Driving Intelligence team)
Leverage premium sensor data and cutting-edge infrastructure to validate algorithms in real-world conditions (Offline Driving Intelligence team)
Collect and generate datasets for specialized vehicle classification and semantic enrichment (Perception Attributes team)
Design and frame machine learning problems for real-world autonomous driving scenarios (Perception Attributes team)
Train and evaluate state-of-the-art machine learning models with a focus on computer vision (Perception Attributes team)
Collaborate with engineers to deploy models for real-time inference on vehicles (Perception Attributes team)
Contribute to improving vehicle's ability to recognize and respond to emergency vehicles, school buses, construction vehicles, and other specialized road actors (Perception Attributes team)
Develop advanced ML models to perceive vehicle's surroundings, identify hazards, and driving restrictions (Perception Scene Understanding team)
Utilize vision-language models for detecting rare events and ensuring safe driving (Perception Scene Understanding team)
Work with state-of-the-art machine learning models that operate in real-time on robotaxi platform with minimal latency (Perception Scene Understanding team)
Collaborate with engineers and researchers across sensors, planning, and other teams to validate algorithms (Perception Scene Understanding team)
Develop multimodal foundation models as common backbone for on-vehicle perception to detect long-tail events and generalize to new geofences (Occupancy and Rare Events team)
Develop effective tokenization techniques for Vision, Lidar, and Radar modalities (Occupancy and Rare Events team)
Leverage LLM techniques to align token embeddings across modalities into a common feature space supporting 3D tasks (detection, segmentation, tracking, feature matching, dense depth) (Occupancy and Rare Events team)
Collaborate with engineers across PCP, MLInfra, and Offboard Driving Intelligence teams, utilizing large-scale dataset to train and evaluate models (Occupancy and Rare Events team)
Build optimized inference pipelines for on-bot algorithms (Perception Optimization team)
Experiment with optimizing SOTA large ML models using techniques such as quantization, pruning, advanced transformer optimizations (token pruning, merging, layer pruning) to deploy on-bot in real-time (Perception Optimization team)
Apply post-training optimization (e.g., quantization) and architectural approaches (e.g., token merging) to fit models into on-bot compute (Perception Optimization team)

Skills

Machine Learning

Computer Vision

Large Language Models

Multimodal Models

Model Fine-tuning

Dataset Generation

Real-time Inference

Autonomous Driving

Sensor Data

Scenario Understanding

Zoox

Develops autonomous electric ride-hailing vehicles

About Zoox

Zoox focuses on creating a fully integrated autonomous ride-hailing service designed specifically for urban transportation. Unlike other companies that modify existing vehicles for self-driving capabilities, Zoox has engineered its own vehicle from the ground up, optimizing it for safety, efficiency, and passenger comfort. The vehicle features advanced sensors, cameras, and AI systems that allow it to navigate complex city environments. It can carry up to four passengers in a spacious interior, ensuring a pleasant ride experience. Safety is paramount, with rigorous testing and collaboration with regulatory bodies to meet high standards. Additionally, Zoox's vehicles are fully electric, promoting sustainability by reducing emissions. The company also develops a user-friendly ride-hailing platform that uses data analytics to improve routing and reduce wait times. Zoox aims to enhance urban mobility while prioritizing safety and environmental responsibility.

Menlo Park, CaliforniaHeadquarters

2014Year Founded

$1,206.2MTotal Funding

ACQUISITIONCompany Stage

Automotive & Transportation, Energy, AI & Machine LearningIndustries

1,001-5,000Employees

Benefits

Health Insurance

Maternity & Paternity Leave

Vacation & Paid Time Off

Sick Days

Free Lunch or Snacks

Employee Assitance Program

Risks

Increased competition from Waymo and Cruise may impact Zoox's market share.

Expansion into new cities introduces regulatory and operational challenges for Zoox.

Partnership with Williams Racing may divert focus from core vehicle development.

Differentiation

Zoox designs vehicles specifically for autonomous operation, unlike retrofitting existing cars.

Zoox's vehicles feature a unique, futuristic design with advanced AI systems.

Zoox offers a fully integrated ride-hailing service optimized for safety and efficiency.

Upsides

Zoox's expansion into Austin and Miami opens new markets and data collection opportunities.

Partnership with Williams Racing boosts Zoox's brand visibility in the U.S. market.

Hiring former Tesla executive Gao could accelerate Zoox's technological advancements.

Land your dream remote job 3x faster with AI

Try Jobo Free