ML Engineer, End-to-End Autonomy

Blue River Technology•Santa Clara, CA

1d•$135,000 - $378,000•Hybrid

About The Position

We’re Blue River, a team of innovators driven to create intelligent machinery that solves monumental problems for our customers. We empower our customers – farmers, construction crews, and foresters - to implement safer and more sustainable solutions, driving increased profitability with less reliance on scarce labor. We believe that focusing on the small stuff – pixel-by-pixel and task-by-task - leads to big gains. With our partners at John Deere, we have the ability to bring innovative computer vision, machine learning, robotics, and product management solutions to scale production, maximizing their potential impact. Our people are at the heart of what we do. Through cross-disciplinary collaboration, this mission-driven and daring team is eager to define the new frontier of mobile robotics. We are always asking hard questions, rapidly iterating, and getting our boots in the field and on-site to figure it out. We won’t give up until we’ve made a tangible and positive impact on the planet. Summary Our mission is to bring the power of E2E machine learning and robotics to John Deere, revolutionizing how robotic systems are built. We aim to open up previously inaccessible opportunities, delivering autonomy products that customers love. As an ML Engineer, you will help grow and shape our E2E stack. This involves shaping the design of the E2E training and inference pipeline, both on- and off-vehicle and on- and off-prem. You will also help with new models and features. Our E2E team is lean and moving quickly. You should be comfortable and excited about working with ambiguity, helping define what will move the program forward, working across traditional boundaries, and learning new things.

Requirements

Proven track record in developing ML models. Prefer experience deploying to production for end-to-end robotics and boosting key metrics.
Expertise in designing and developing software for complex systems.
Comfortable working on new hardware systems and working on new ML/software problems.
Strong Python coding skills and proficiency with deep learning frameworks like PyTorch.
Comfortable working across traditional team boundaries to deliver results.
Excellent brainstorming, creative thinking, mathematical analysis, and communication skills.
Track record of regularly anticipating technical issues and making architectural and design decisions to avoid them.

Nice To Haves

Experience with robotics middleware such as ROS or other robotics-focused software packages.
Strong CUDA background or other GPU frameworks.
Experience in multidisciplinary environments. (We’ve got CS, CV/ML, EE, ME, etc.)
Have worked on embedded systems.
Experience with system architecture.

Responsibilities

Design and implement end-to-end policies for complex navigation and manipulation tasks with long horizons.
Enhance policy robustness to challenges, including environmental variability, machine wear, and deployment across machine forms.
Drive the full data + modeling life cycle, from data collection requirements, experimental design, evaluation frameworks, model training, and deployment. Iteratively drive performance improvements.
Develop pretraining strategies leveraging multimodal data at scale. Explore approaches to leveraging John Deere's vast store of machine logs, and bring an E2E lens to inform future logging vision.
Edge deployment: help deliver performance with constrained compute.
Design and implement systems for active learning on the edge.
Collaborate with infrastructure engineers on scalable data pipelines, including ingest, curation, and processing training artifacts.
Collaborate with infrastructure engineers on scalable distributed training pipelines.
Collaborate with reinforcement learning, robotics, and hardware engineers to integrate ML seamlessly into the full E2E autonomy stack.
Work with product management to learn from customers, including how they want to interface with the product. Translate this to ML architecture and implement.
Contribute to the design, development, and validation of perception systems for safety.
Help shape the theory of operations and build the safety testing framework.
Drive the E2E vision and act as an ambassador for an E2E-first organization.
Technical leadership and mentorship: guide the team through complex algorithmic tradeoffs, elevate our approach to empirical testing, and mentor the next wave of E2E robotics leaders.
Stay up to date with the latest research and integrate advancements into our stack.