Software Engineer, Numerical Computing

Description

The company builds defense products on a multi-product portfolio, and its systems run on a platform called Coherent Distributed Networks. Its customers include warfighters, commercial air operators, and border protection teams who need to respond quickly and keep pace with changing threats.

Founded in 2022, the company has raised $1 billion from investors including 8VC, Accel, and Valor Equity Partners. It is headquartered in Los Angeles and has offices in Washington, D.C., San Francisco, San Diego, and London. The company has over 500 employees.

The Spectrum Sensing team is turning radar systems into distributed sensing systems that detect, classify, and geolocate RF emissions. The team's algorithms are only useful if they run correctly, fast, and reliably in production.

This is a software engineering role first. You will decide how the team's numerical and signal processing code is structured, integrated, tested, and sped up. No RF or radar background is required. You do need to diagnose why Python code is slow, know how to fix it, and build code that other engineers can change safely.

You will work with signal processing and RF engineers who create the algorithms, and turn their work into dependable software.

Responsibilities

• Shape the software architecture and engineering practices behind the sensing stack
• Build and maintain production-grade numerical software in Python, with clear interfaces, composable components, and code that stays maintainable over time
• Connect third-party software and hardware to the platform through APIs, and design the abstractions that stop those integrations from spreading through the codebase
• Profile signal processing code and speed it up using vectorization, array frameworks, JIT compilation, GPU offload, or bindings to compiled code, then measure the gains
• Set the testing and CI standards for numerical code, including regression tests, tolerance and reproducibility plans, property-based and reference-implementation tests, and automation that finds breakage before a field demo does
• Work with signal processing engineers to convert algorithms from papers, prototypes, and MATLAB into shipped software
• Help make architecture decisions for the sensing product line, including how components fit together across the platform
• Travel periodically, up to 30% of the time

Requirements

• Bachelor's degree in computer science, electrical engineering, applied mathematics, physics, or a related field
• 5+ years of professional software development experience
• Experience writing, maintaining, testing, and debugging production Python code for numerical work, beyond analysis scripts and notebooks
• Strong working knowledge of array-based numerical libraries such as NumPy, SciPy, or equivalents, and an understanding of where Python slows down and why
• Experience speeding up numerical code with at least one of: array or tensor DSLs and JIT compilers (JAX, Numba, PyTorch, Triton, Halide, or similar), GPU computing (CUDA or equivalent), or Python bindings to compiled languages (pybind11, nanobind, Cython, or similar)
• Experience working in a large codebase shared by several teams, including working within existing architecture, reading unfamiliar code, and shipping changes other teams rely on
• Solid engineering habits: version control workflow, code review, automated testing, and CI
• Clear written and verbal communication, including documenting technical details and presenting findings
• Must work on-site four days per week in Washington, D.C.
• U.S. citizen
• Able to obtain and maintain a U.S. security clearance
• Preferred: master's degree in computer science, electrical engineering, applied mathematics, or a related field
• Preferred: meaningful open-source contributions to numerical, scientific computing, or signal processing software, judged from commit history rather than repository lists
• Preferred: exposure to digital signal processing such as filtering, FFT-based processing, detection and estimation, or beamforming, from coursework, research, or industry
• Preferred: experience with software-defined radios or related toolkits such as GNU Radio
• Preferred: experience with real-time, streaming, or throughput-limited data processing
• Preferred: experience designing APIs or libraries used by other engineering teams
• Preferred: familiarity with MATLAB, enough to read and port existing algorithm prototypes
• Preferred: current U.S. security clearance
• Preferred: experience with agile tools such as JIRA and Confluence
• Preferred: experience in a DoD, military, or other government setting with fast-moving, changing priorities

Conditions

Work is on-site in Washington, D.C., four days per week

The posted salary range covers base pay only and excludes bonus, equity, and benefits

The range is an estimate, and the actual offer may vary based on experience, education or training, critical skills, and business needs

Candidates whose pay expectations fall outside the range are still encouraged to apply

Benefits

• Medical, dental, and vision coverage, with 100% of premiums paid by the company
• 401k with a 50% company match on contributions up to 6% of pay
• FSA, HSA, life insurance, and more
• Free lunch every workday when onsite
• Fully stocked kitchen
• Unlimited PTO
• Casual dress code
• Competitive base salary, generous pre-IPO stock option grants, and relocation assistance
• Annual bonuses, coming soon