Lead Learning-Driven Autonomy Controls Engineer

Teleo; Palo Alto (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

Teleo, a Havoc company, is building autonomous heavy equipment robots for commercial and defense applications. You will join a team engineering learning-driven control policies that replace manually tuned MPC, enabling a single operator to supervise multiple machines while preserving safety and interpretability.

In this role you will own the transition, develop data-driven plant models, test plans, and interfaces between classical control and learning components, and validate policies on real

Qualifications

• 2-3 years of experience with experimental data collection and data analysis to estimate parameters of a plant model for automatic control.
• Proficient in production-quality C, C++, or Python code.
• Strong understanding of modern robotics control systems.
• Experience with learning-based control or policy optimization for real-world systems.
• Ability to work near hardware with real-time constraints.

Responsibilities

• Transition from MPC-based to learning-driven control policies with cross-vehicle adaptability.
• Collaborate with Principal Controls Engineer to translate control insights into learning-friendly formulations.
• Define interfaces between classical control (MPC, PID, state estimation) and learning components.
• Validate learned control policies before real-vehicle deployment.
• Develop test plans and collect real-world data for model identification and validation.
• Integrate learned controllers into the existing vehicle control stack safely.

Skills

Data collection & analysis

C/C++/Python

Robotics control systems

Learning-based control

Real-time constraints

Job description

Teleo, a Havoc company, is building autonomous heavy equipment robots for commercial and defense applications. You will join a team engineering learning-driven control policies that replace manually tuned MPC, enabling a single operator to supervise multiple machines while preserving safety and interpretability.

In this role you will own the transition, develop data-driven plant models, test plans, and interfaces between classical control and learning components, and validate policies on real