Machine Learning Research Engineer

About WindBorne Systems

WindBorne Systems enhances weather predictions through a unique global network of advanced smart weather balloons that gather crucial atmospheric insights. The company designs, builds, and operates these balloons, generating weather data that was previously unattainable. Their goal is to remove uncertainty from weather forecasts and support humanity's adaptation to climate change, from forecasting hurricanes to facilitating renewable energy adoption. Founded by Stanford engineers recognized by Forbes 30 Under 30 in 2019 and backed by premier investors including Khosla Ventures and Footwork VC, WindBorne runs AI-driven weather models non-stop, delivering worldwide forecasts every 20 minutes.

Key Responsibilities

• Develop AI-driven data assimilation methods to integrate real-time observations from diverse sources such as balloons, satellites, and weather stations into weather forecasts.
• Create a comprehensive foundation model aimed at predicting diverse weather-related datasets, including variables beyond conventional global forecasts.
• Research, clean, and merge extensive, heterogeneous weather data sets with varying formats, resolutions, and quality; identify valuable data and build systems for efficient reuse.
• Conduct rapid experimentation cycles to test hypotheses, learn from failures, and drive incremental model performance improvements.
• Convert successful research experiments into reusable infrastructures that facilitate and accelerate future work.
• Shape research directions by hypothesizing, designing rigorous experiments, staying updated on relevant ML advancements, and prioritizing promising ideas.
• Engage in various technical tasks beyond core research including operations, infrastructure development, evaluation, and data engineering as required to deploy research effectively.

Required Qualifications

• Strong research acumen with the ability to identify significant questions, design conclusive experiments, and validate findings.
• Passion for machine learning with an in-depth understanding of models beyond treating them as black boxes.
• Proficiency in Python and PyTorch, experienced in developing and debugging sophisticated ML systems.
• Expertise handling large, unstructured datasets and building robust pipelines or abstractions.
• Ability to rapidly iterate while discerning which components should be established as durable infrastructure.
• Comfortable operating at the intersection of research and engineering and capable of learning new tools or domains as necessary.
• Experience or knowledge in weather, climate, geospatial data, scientific machine learning, computer vision, or time-series forecasting is advantageous but not mandatory.

Employee Benefits

• 401(k) retirement plan options
• Comprehensive dental, health, and vision insurance
• Unlimited paid time off
• Stock option scheme participation
• Office-provided food and beverages

Compensation & Location

The role offers an annual salary range of $140,000 to $240,000, tailored to candidate background and market competitiveness. Work is based at 1600 Bridge Parkway, Redwood City, CA, on a hybrid or in-person schedule.