Senior Dataset Owner - Robotics

At SWARM Biotactics, we’re creating biorobots — living, intelligent systems based on insects. Each insect carries a custom-designed backpack enabling sensing, wireless control, and secure communication.

Our systems rely on large amounts of heterogeneous, real-world sensor data tightly coupled with autonomy, perception, and learning.

To scale our technology, data is a core system, not a byproduct.

We are looking for a senior engineer who takes full ownership of dataset generation and management — from defining what data is required, to collecting it in the real world, to making it usable, traceable, and reliable for downstream algorithms.

🚀 Join us at the frontier — and build the data foundation for the first generation of living machines.

💡 Your Role — Data & Dataset Ownership

You will own the entire lifecycle of data used in our robotic systems.

Your work will directly determine the quality, robustness, and scalability of our perception, learning, and autonomy stacks.

This role is about ownership, structure, and intent — not just processing existing data.

What you’ll do

• Take end-to-end ownership of dataset generation for real-world autonomous systems
• Define what data is needed, when it is needed, and why
• Plan and execute targeted data collection campaigns aligned with development and validation goals
• Define annotation concepts and formats that reflect downstream use cases
• Build and maintain data preprocessing and validation pipelines
• Design robust dataset structures, metadata schemas, and versioning strategies
• Ensure traceability, reproducibility, and long-term usability of datasets
• Act as the interface between data acquisition and algorithm development
• Establish best practices for data quality, documentation, and usage across teams

🧠 Your Profile

• Master’s degree (or equivalent experience) in Robotics, Computer Science, Electrical Engineering, or a related field
• 3+ years of experience working with real-world sensor data for perception, localization, or environment understanding in robotic or autonomous systems
• Solid understanding of inertial sensing, motion estimation, and 2D/3D environmental perception, including typical noise characteristics, failure modes, and calibration aspects
• Experience designing data pipelines and dataset abstractions, not just consuming datasets
• Hands-on experience with cloud-based data infrastructure
• Experience with AWS fundamentals for data and ML workflows
• Experience with Amazon S3 for dataset storage, structuring, and lifecycle management
• Experience with Amazon SageMaker for dataset usage, experimentation, or integration with ML pipelines
• Experience with AWS Glue, AWS Athena
• Demonstrated ability to take long-term technical ownership of complex, cross-disciplinary systems
• Structured, analytical mindset with a strong sense for data quality and failure modes
• Ability to understand how data is used by perception, learning, and autonomy components

Nice to Have / Plus

• Experience with robotics, autonomous systems, or embedded platforms
• Familiarity with dataset versioning, experiment tracking, and reproducibility
• Experience supporting learning-based or data-driven pipelines
• Exposure to field testing or real-world experimentation
• Experience defining access control, data governance, or cost-aware data management in cloud environments

🌍 Why SWARM Biotactics

• We’re building technology that has never existed before
• You’ll define foundational data systems where no standards exist yet
• High ownership and direct impact on core product capabilities
• Close collaboration with experts across biology, hardware, and AI
• A role with real responsibility — not a “support” position

⚡ Join the Swarm

If you want to own data as a first-class system and shape how real-world autonomous systems learn and improve, we’d love to hear from you.

✨ Application note

Please include a concrete example where you took end-to-end responsibility for data or datasets, including the key decisions you made and the trade-offs involved.