Senior Research Engineer, Controls

present of AI; Santa Clara (CA); On-site; USD 140,000 - 230,000; Full time

Job summary

PlusAI in Santa Clara is building the software that turns planned driving paths into real steering, throttle and braking for heavy trucks. You’ll ground control policies in solid theory and work with large real-world datasets to learn vehicle behavior.

The role blends research with hands-on development, focusing on data quality, model validation and safety-aware shipping of learned components within the autonomy stack.

Qualifications

• Strong foundation in control theory and motion planning.
• Hands-on experience building ML models of vehicle behavior.
• Experience with large real-world driving datasets for learned policies.

Responsibilities

• Develop and test learned vehicle models and control policies.
• Analyze vehicle performance and tune systems on real trucks.
• Collaborate with research scientists to integrate learned components into safety-critical software.

Skills

Control theory

Motion planning

Machine learning

Vehicle data

Tools

C++

Python

Job description

The role centers on the software that turns a planned driving path into actual steering, throttle and braking on a heavy truck. A strong candidate has solid control theory and motion planning grounding, plus hands-on experience building machine-learning models of vehicle behavior and learned control policies from large real-world datasets. Comfort with analyzing vehicle performance, tuning and tracking down problems on real trucks matters as much as research skill.

How it fits the AI era

AI is moving into the core of this work, since vehicle models and control policies are learned from fleet data instead of only being hand-derived. That shifts effort toward data quality, model validation and knowing when a learned approach beats a classical one. Judgment about safety, reading odd behavior on a real vehicle and deciding what is good enough to ship stay human.

How it builds AI experience

The work offers direct practice training and testing learned vehicle models and control policies on data from a working autonomous trucking fleet. Someone preparing for it could build a small learned dynamics model from public driving or vehicle data and compare it against a classical controller in simulation. Working beside research scientists on a truck-focused autonomy stack adds exposure to how learned parts fit with safety-critical software.

Worth showing

• Learned vehicle dynamics models
• Reinforcement or imitation learning policies
• Vehicle data analysis and tuning
• C++ or Python for autonomy

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

PlusAI develops AI-based virtual driver software for factory-built autonomous trucks, per the company.