Staff Software Engineer, IR Data Products

Description

The company builds and runs a growing set of infrared (IR) space missions that produce Earth observation data for many customers. The FireSat constellation delivers calibrated IR imagery and derived wildfire products to wildfire operators, the scientific community, the US Government, and commercial partners. The Space Based Environmental Monitoring (SBEM) mission supplies data for weather and cloud characterization, situational awareness, and other use cases.

The IR Data Products team owns the multi-mission, multi-customer software that processes, stores, and distributes this data. That covers turning raw downlink into calibrated imagery and derived products, then delivering them reliably, with low latency, at scale.

This is a staff-level architecture and technical leadership role within the Software focus area of the team. You will turn longer-term technical direction into multi-quarter designs and production systems, leading initiatives that make the system faster, cheaper, more reliable, easier to evolve, and ready for many satellites.

Your work will span four areas. First, you will shape the next-year architecture for pre-processing, processing directives and routing, delivery orchestration, compliance evaluation, and the orchestration platform. Second, you will guide the low-latency roadmap, including event-driven microservices, cloud infrastructure, and algorithm work. Third, you will set the direction for storing, cataloging, serving, and versioning geospatial data such as NetCDF, Zarr, HDF, STAC, and tile stores. Fourth, you will build the production base for ML products like model packaging, experiment tracking, and GPU training and inference, which the team's hotspot, perimeter, and fire behavior products will increasingly rely on.

You will also represent the team in cross-functional design reviews, lead prioritization with Technical Program Managers and Business Development, set software standards, and mentor senior and mid-level engineers on failure modes, immutability, idempotency, and reprocessing.

The ideal person is a hands-on staff engineer who has led multi-quarter architecture work on production data platforms at scale. They know distributed data systems and modern workflow orchestration, have real geospatial experience and ideally MLOps depth, and can write a design document and align teams as well as debug difficult algorithm code.

The role is hybrid, with on-site work at the Denver, CO office three days per week.

Responsibilities

• Co-own the architecture of the multi-mission, multi-customer IR data systems with the Principal Software Engineer, covering ingest, pre-processing, orchestration, storage and catalog, delivery, and the platforms that run them.
• Lead critical capabilities for the IR missions from problem framing and design document through rollout, adoption, and operational handoff.
• Own key parts of the low-latency roadmap, including event-driven microservices, cloud infrastructure design and optimization, and algorithmic optimization.
• Set direction for geospatial storage, cataloging, and serving at scale, weighing formats, tile stores, STAC catalogs, bucket and IAM layout, and reprocessing and versioning strategy.
• Design and build MLOps foundations for derived IR products: model packaging and versioning, experiment tracking, GPU training and inference, A/B evaluation harnesses, and reproducibility, so ML products can move from research to production smoothly.
• Set team standards for code organization, release engineering, CI, testing, documentation, code review, observability, and operations, and raise the team's skill at reasoning about failure modes, immutability, idempotency, and reprocessing.
• Lead technical work across functions with Product, Program Management, Business Development, and other Software teams. Represent IR Data Products in design reviews and turn unclear requirements, such as latency budgets, compliance constraints, new missions, and new customers, into concrete engineering plans.

Requirements

• Bachelor's degree with 10+ years, MS with 8+ years, or PhD with 5+ years of software engineering experience.
• Proven technical leadership on production data platforms at scale, with designs and rollouts to show.
• Experience designing and running production data pipelines on AWS with a modern workflow orchestrator such as Flyte, Airflow, Prefect, Argo, or similar.
• History of leading multi-quarter architecture initiatives across several services or teams.
• Experience reducing latency through measurement, benchmarking, per-stage instrumentation, right-sizing, fan-out, caching, and async I/O, and recognizing when a fix must be structural rather than algorithmic.
• Strong operational habits: observability, SLOs, on-call, safe deploys, hotfix protocols, immutability, and reproducibility.
• Strong technical writing and cross-functional communication, able to align engineers, scientists, PMs, and adjacent platform and flight-software teams.
• Preferred qualifications: experience with event-driven or microservices architectures, including queues, event routing, backpressure, idempotency, retries, DLQs, and clear service contracts.
• Preferred qualifications: familiarity with modern workflow platforms such as Union.ai or Flyte v2.
• Preferred qualifications: Terraform or IaC ownership of AWS resources, and hands-on work shaping IAM boundaries between environments.
• Preferred qualifications: prior work on geospatial or remote sensing systems at scale.
• Preferred qualifications: strong grasp of the tradeoffs among geospatial storage formats, tile stores, and STAC.
• Preferred qualifications: hands-on MLOps experience with model packaging and versioning, experiment tracking such as MLflow or W&B, GPU training and inference, feature stores, and A/B harnesses.
• Preferred qualifications: experience with GPU acceleration for image processing, such as JAX, CuPy, or CUDA, or with ML inference at scale.
• Preferred qualifications: experience in satellite ground segment or a similar domain where compliance, provenance, and data lineage matter.
• Preferred qualifications: experience introducing latency SLOs and benchmarking discipline to a team.
• Preferred qualifications: prior experience serving imagery or tiled geospatial data to external customers or partner platforms.

Conditions

The base salary range is $216,000 to $239,000, plus an equity grant and a full benefits package. Final pay depends on skills, qualifications, experience, and location, as assessed during interviews.

Hybrid role, on-site in Denver, CO three days per week.

The company is registered to hire in AL, AZ, CA, CO, CT, DC, FL, GA, IL, IA, ID, MD, MA, MI, MN, MO, NJ, NY, NC, OH, OR, PA, TN, TX, VT, VA, and WA.

The role requires access to export-controlled information. Applicants must be a U.S. person (U.S. citizen or national, lawful permanent resident, refugee under 8 U.S.C. § 1157, or asylee under 8 U.S.C. § 1158), be eligible to access such information without a required export authorization, or be eligible and reasonably likely to obtain one. The company may decline to pursue export licensing for legitimate business reasons.

Benefits

• Equity compensation for full-time employees.
• Medical, dental, and vision insurance, a 401k plan, short- and long-term disability, and life insurance.
• Three weeks of paid vacation for new employees, 12 paid holidays, unlimited sick time, and paid parental leave.