Lead AI/ML Engineer Engineering · San Francisco, CA (Hybrid) ·

Engineering · San Francisco, CA (Hybrid) · Full time · Posted 5 months ago

About Vega

Vega is a pre-seed enterprise SaaS company in the aerospace and defense sector. We provide spectrum orchestration software to satellite operators, helping them forecast interference and exchange spectrum. Customers include commercial satellite operators, civil agencies and regulators, and the defense community.

The role

We're looking for a Lead AI/ML Engineer to drive the development of machine learning models that predict and analyze RF spectrum interference in the space domain. You'll work at the intersection of orbital mechanics, signal processing, and modern ML techniques to build models that help satellite operators understand and mitigate interference risks. This is a founding engineering role — you'll shape our ML infrastructure and model architecture from the ground up.

What you'll do

• Design and implement ML models for satellite coverage prediction and RF interference analysis
• Build data pipelines that process orbital ephemeris data (TLEs, SP3) and spectrum allocation data
• Develop time-series models for predicting interference patterns across HEALPix coordinate grids
• Create training datasets from simulation outputs and real-world spectrum monitoring data
• Establish MLOps practices including model versioning, automated testing, and deployment pipelines
• Collaborate with domain experts on orbital mechanics and RF propagation modeling
• Optimize model inference for production workloads handling millions of satellite-epoch combinations

Requirements

• 5+ years of experience in machine learning engineering or applied ML research
• Strong proficiency in Python and ML frameworks (PyTorch, TensorFlow, JAX)
• Experience with time-series forecasting or spatial/geospatial ML applications
• Hands-on experience with MLOps tools (MLflow, Kubeflow, or similar)
• Familiarity with distributed computing and cloud platforms (AWS, GCP)
• Experience deploying ML models to production at scale
• Strong software engineering fundamentals (testing, version control, code review)

Nice to have

• Background in aerospace, satellite communications, or RF engineering
• Experience with orbital mechanics libraries (Orekit, SGP4, GMAT)
• Familiarity with HEALPix or other hierarchical spatial indexing schemes
• Experience with graph neural networks or attention mechanisms for spatial data
• Knowledge of ITU spectrum regulations and satellite frequency coordination
• Experience with GPU programming and CUDA optimization

Questions?

Reach out to our team if you have questions about this role: View email address on click.appcast.io .