Software Engineer, Lodestar
On-board Autonomy
$140-180k
+ Equity incentives
More information about location
Space research technology
Open for applications
Space research technology
1-20 employees
Open for applications
$140-180k
+ Equity incentives
More information about location
1-20 employees
Company mission
To deliver orbital defence as a service to high-value US and allied assets.
Role
Who you are
- Bachelor’s or Master’s degree in Computer Science, Machine Learning, Robotics, a related field, or equivalent experience
- 2+ years of distinguished industry experience in autonomy, decision-making, or control systems for aerospace/robotics
- Strong proficiency in C++ and Python and DL frameworks (PyTorch, TensorFlow)
- Demonstrated experience with machine learning applied to decision-making or control problems
- Track record with optimal control, planning, or reinforcement learning in real-time systems
- Familiarity with multi-agent decision-making or planning under uncertainty
Desirable
- Track record implementing autonomy applications in real-time or safety-critical environments
- Experience integrating perception and prediction outputs into decision frameworks
- Familiarity with resource-aware strategy selection and optimization under uncertainty
- Background in reinforcement learning, hierarchical planning, or adaptive control
- Experience with distributed training and cloud-based scaling of ML models (AWS, GCP, or Azure)
- Experience with Linux, Git, and CI/CD pipelines
- Comfortable with containerization tools such as Docker and Kubernetes
- Familiarity with real-time systems, multi-threading, and performance optimization
- Strong understanding of distributed autonomy, networking, and communication protocols
What the job involves
- At Lodestar, as a Software Engineer - On-board Autonomy, you’ll be developing the decision-making systems at the heart of Lodestar’s autonomy suite
- You’ll focus on creating algorithms that evaluate mission context, system capabilities, and environmental conditions to recommend and execute optimal actions
- These models adapt dynamically to changing conditions, enabling real-time autonomous decision-making even in communications-limited or uncertain environments
- Your work will bridge perception, prediction, and control, allowing spacecraft to operate intelligently, efficiently, and resiliently across complex mission scenarios
- Design and implement on-board decision-making models that recommend and adapt strategies in real time
- Develop autonomous decision algorithms that integrate information from perception, state estimation, and intent prediction models to execute mission objectives
- Research and implement ML models for decision making - everything from lit. review, through training, to deployment
- Implement decision models that adapt dynamically to changing mission context, environmental conditions, and system status
- Develop frameworks for continuous re-evaluation of active strategies to ensure resilient and adaptive behavior under uncertainty
- Support real-time autonomy in communications-limited or time-critical scenarios
- Build and maintain autonomy infrastructure, testing frameworks, and deployment pipelines for space missions
Our take
Satellites are critical for services ranging from GPS to disaster coordination. But they face significant vulnerabilities due to their inaccessibility, making them susceptible to technical issues and attacks. Addressing these concerns is essential for maintaining space infrastructure and national security.
Lodestar aims to revolutionise satellite maintenance and security with a "first responder" service in space, utilising advanced robotic sentries. Unlike competitors building entire spacecraft, Lodestar focuses on a platform-agnostic robotic system. Its initial product, a dexterous robotic arm, is designed to inspect and repair assets in orbit, offering scalable solutions for commercial and defence customers.
The company has continued to advance it s autonomous in-orbit capabilities, including Mithril, a sensing system using LiDAR, machine vision, adn AI to detec and respond to nearby spacecraft, with flight demonstrations planned for 2026. The company has also partnered with Momentus to integrate its robotic arm on partner spacecraft, expanding applications across commercial and defense missions. These developments reinforce Lodestar's role in delivering scalable, autonomous solutions for satellite inspection, maintenance, and security.
Kirsty
Company Specialist at Welcome to the Jungle
Company
Funding (2 rounds)
Dec 2025
$0.1m
GRANT
Jul 2024
$2.5m
SEED
Total funding: $2.6m
Company HQ
SoMa, San Francisco, CA
Leadership
Neil Buchanan
(Co-Founder)Graduated from the University of Edinburgh with a Master's degree in Mechanical Engineering. Previously founded endeavour where they served as Team Lead for 2 years. They also worked as a Business Development Specialist at Flowcopter.
Thomas Santini
(Co-Founder)Previously Co-founded Dekked. They also worked as a Product Manager at Backslash Build and Nutribloc, and as an RDD New Product Design Engineer at Dyson.
Salary benchmarks
We don't have enough data yet to provide salary benchmarks for this role.
Submit your salary to help other candidates with crowdsourced salary estimates.
Share this job
View 2 more jobs at Lodestar