Applied AI/ML Engineer

About The Role

At Ursa Space Systems, we transform complex satellite and spatial data into actionable insights through our GeoAI platform, which lets users ask questions in simple English about any location or event on Earth. The system manages data selection, sensor tasking, analysis, and generates reports in minutes, accelerating decision-making.

We are seeking an Applied AI/ML Engineer to develop the intelligent systems driving this platform. This position involves varied responsibilities, alternating between developing computer vision models on SAR and electro-optical satellite imagery, and creating agentic AI workflows using large language models (LLMs). The candidate will encounter a dynamic project environment that embraces diversity in tasks.

Reporting to the Director of Analytics, the role works collaboratively with data scientists, image scientists, product owners, and customers. The role requires occasional on-call support during evenings or weekends. This is a fully remote, exempt role.

Job Summary

• Design and implement agentic AI systems that automate geospatial analysis, from natural language query processing to multi-source data analysis and report production.
• Develop orchestration and context-management features connecting LLMs with geospatial data and analytics across 90+ integrated data feeds.
• Perform verification and validation of AI/ML and LLM outputs to ensure result accuracy.
• Train, fine-tune, evaluate, and deploy computer vision models (object detection, segmentation, change detection) on Earth observation datasets including SAR and electro-optical imagery.
• Adapt vision-language models for comprehensive scene understanding beyond object localization.
• Build infrastructure to evaluate classical machine learning and LLM agent performance and use metrics to guide improvements.
• Manage projects end-to-end including data preparation, prototyping, deployment, validation, and maintenance.
• Collaborate with product teams and customers to translate requirements into system designs and communicate technology functionalities.
• Integrate diverse third-party and multi-source datasets into analytic pipelines.
• Conduct ad-hoc analyses responding rapidly to organizational inquiries.
• Serve as a technical mentor within the team by sharing knowledge of new AI models and technologies.
• Measure and monitor actual system errors versus allocated performance budgets.
• Perform adversarial and edge-case testing to identify potential failure modes in ambiguous or conflicting data.
• Calibrate system confidence scores to correspond with real-world accuracy rates.
• Ensure traceability and reproducibility to explain specific outputs based on input data and model versions.

Requirements

• Bachelor’s or Master’s degree in Computer Science, Data Science, or related STEM fields.
• 4 to 6 years’ experience developing and deploying machine learning solutions in software or product-focused environments.
• Proficiency in Python with strong software engineering practices including Git, Docker, testing, code review, and CI/CD.
• Experience training and deploying deep learning models for computer vision tasks using frameworks such as PyTorch.
• Hands-on background in building LLM-driven applications incorporating prompt engineering, structured outputs, tool use, and agentic architectures.
• Expertise in ML system evaluation using metrics like RMSE, FPR/TPR, AUC/ROC, IoU, and methods for LLM/agent performance assessments.
• Practical experience deploying in AWS cloud environments.
• Strong communication skills capable of explaining complex models and results to both technical and non-technical audiences including customers.

Preferred Qualifications

• Prior work with remote sensing data, including SAR and GIS spatial statistics.
• Experience fine-tuning foundation or vision-language models with techniques such as LoRA/PEFT targeting domain-specific imagery.
• Knowledge of agent orchestration frameworks like LangGraph, Claude Agent SDK, MCP, and LLM evaluation tools.
• Familiarity with geospatial Python libraries such as GDAL, rasterio, geopandas, and xarray.
• Experience with SQL/NoSQL databases and vector stores.
• Leadership experience managing complex, interdisciplinary projects.
• Working knowledge of collaboration platforms such as Confluence, Jira, and Miro.
• Expertise in regression and model drift detection to identify subtle degradations in ML performance.

Compensation

• Annual salary range is $185,000 to $199,000, inclusive of base pay and eligibility for annual bonuses.
• Starting salaries typically fall between the range minimum and midpoint, influenced by skills and experience evaluated during hiring.

Inclusion Statement

We are committed to diversity, equity, and inclusion, supporting a workplace that values all voices and experiences. Employment decisions are made without regard to race, color, age, religion, gender, sexual orientation, disability, veteran status, and other protected characteristics.

Location

Our headquarters is in Ithaca, NY, with a distributed remote workforce throughout the United States.

Application Note

Submitting a relevant cover letter expressing genuine interest in Ursa Space Systems is mandatory for consideration.

Benefits and Perks

• Competitive salary and discretionary paid time off with flexible scheduling.
• Stock option plans and 401(k) matching.
• Comprehensive health coverage including medical, dental, and vision for employees and dependents.
• Flexible spending and health savings accounts.
• Employer-provided life insurance, short-term and long-term disability coverage for parental and family care.
• 11 paid holidays annually.
• Employee resource groups fostering community and support.
• Programs for educational assistance and professional development.
• Additional benefits and opportunities to grow professionally.

Company Values

• Leverage teamwork and collaboration.
• Take ownership and solve problems proactively.
• Strive for elegant, simple solutions.
• Promote diversity and inclusivity.
• Act ethically and responsibly.
• Be resourceful and innovative.