Senior Data Scientist

Who you are

• BS/MS/PhD degree in Computer Science, Statistics, Applied Mathematics, Operations Research, Aerospace Engineering, or a related quantitative field
• 2+ years of applied data science or machine learning experience, including models that have run in production
• Strong Python and strong SQL over large analytical datasets
• Depth in time-series and sequence modeling
• Experience working with sparse, incomplete, or unlabeled data — weak supervision, programmatic labeling, and building evaluation sets where no ground truth exists
• Experience designing data validation and drift detection for continuously arriving data
• Able to leverage AI assistants to increase velocity while fully understanding and taking absolute ownership of every model, feature, and analysis you ship
• Strong communication skills and a collaborative mindset, including the ability to explain methodology and its limitations to a non-specialist audience
• Prior experience or a deep interest in aerospace, aviation, or high-throughput tracking systems — including familiarity with flight telemetry, ATC phraseology, or airspace procedures
• Familiarity with geographic/spatial data processing — projections, spatial indexing, and great-circle geometry
• Awareness of distributed messaging and stream-processing systems (i.e. Apache Pulsar or Kafka, Apache Flink) and columnar or lakehouse analytical storage

What the job involves

• As a Senior Data Scientist here, you’ll be the bridge between raw, messy aviation data and the intelligence that powers our AI platform. You aren’t just building models; you’re taking high-frequency aircraft telemetry and noisy air traffic control audio and figuring out what's actually happening in the sky
• Derive new signals from high-throughput, real-time data streams, inferring complex physical and operational states that are not directly observable in the raw telemetry
• Build, evaluate, and iterate on models across diverse modalities, including dense time-series, spatial data, and highly noisy, unstructured sequential data
• Design evaluation for problems with little or no labeled ground truth, using weak supervision, and programmatic labeling — and defend those definitions to engineers and domain experts
• Partner with backend and platform engineers to take models from offline analysis into production streaming pipelines, and own their behavior once they are live
• Work directly with aviation subject-matter experts to turn operational knowledge into features, labels, and evaluation criteria
• Please note that this job description is intended to provide a general overview of the position and does not include an exhaustive list of responsibilities and qualifications