Machine Learning Engineer

Tally BV; New York (NY); Hybrid; USD 140,000 - 160,000; Full time

Job summary

Floodbase is seeking an experienced Machine Learning Engineer to advance a physics-informed flood model that fuses satellite imagery, rainfall, and ground sensor data. The role emphasizes building robust AI systems from data selection to deployment, with responsibility for validation, monitoring, and reproducibility in production.

You will work with a team of scientists and engineers to push the frontier of flood risk analytics, contributing to emergency response tools and parametric insurance

Qualifications

• Bachelor's in a quantitative field with 4+ years of industry experience, or a graduate degree with 1-2+ years.
• Experience training deep learning models in PyTorch or equivalent, including production deployment.
• Strong Python coding skills and ability to contribute to a shared repo.
• Fundamentals of scientific data analysis and statistics; ability to quantify uncertainty and calibrate models.

Responsibilities

• Develop and maintain observation-driven flood models linking satellite data, weather data, and ground sensors.
• Estimate flood risk and uncertainty for parametric insurance products in collaboration with cross-functional teams.
• Deploy and monitor models in cloud infrastructure, ensure reproducibility and provenance in production.
• Evaluate and integrate advances in GeoAI and geospatial data into the Floodbase product.

Skills

Python

PyTorch

Statistical analysis

Uncertainty quantification

Data validation

Interdisciplinary collaboration

Education

Bachelor's in a quantitative discipline

Graduate degree in a quantitative discipline

Tools

rasterio

GDAL

Shapely

Geopandas

Xarray

Dask

Zarr

Job description

Our observation-driven, physics-informed AI model, built on over a decade of peer-reviewed science, continuously monitors flooding globally and now also powers flood event response for governments, with forecasting and risk analytics launching in 2027. Floodbase is backed by investors including Lowercarbon Capital, Collaborative Fund, and Floating Point, and trusted by NASA, The UN, Google, FEMA, and more.

About the role:

We are looking for an experienced Machine Learning Engineer to advance our observation-driven, physics-informed flood model: a deep learning estimator that uses AI as the connective tissue between satellite imagery, rainfall and weather data, physics-derived state variables, and ground sensors to produce a continuous estimate of flood state across the past, present, and near future. We produce flood data that informs emergency response and enables parametric flood insurance for governments, farmers, and businesses. You will work on the frontier of new capabilities at Floodbase, building the core AI models that enable flood risk transfer through parametric insurance. You will take ownership of large projects, and your work will shape the full pipeline from training data selection to model training, validation, and output transformation and analysis. You will own models through deployment: what you build runs in our monitoring pipelines, and you are responsible for its evaluation, monitoring, and reproducibility in production. You will work with a team of scientists and engineers with expertise in remote sensing, hydrology, index insurance, climate, social vulnerability, and machine learning to build the next generation of tools to ensure financial protection from floods. We produce cutting-edge AI climate science, and you will play a large role in leading the edge of this field, with opportunities to publish in leading journals and present at top tech and science conferences. The role is remote, with core hours in US Eastern Time or hybrid based in New York City, NY or Toronto, ON. Candidates must be authorized to work in the United States or Canada.

What you’ll do:

- Use your analytical expertise to derive robust insights from large structured and unstructured datasets of observations, hydrologic model outputs, and auxiliary records (stream gauges, water occurrence, …) that can be incorporated into the Floodbase product. Example project: Automate the manual validation process of our historical flood data for individual areas of interest (AOIs) by finding robust rules leveraging multiple validation data sources.

- Utilize your proficiency in Machine Learning to estimate risk of flooding and its associated uncertainty bounds informed by observations of flooding and hydrological models for parametric insurance product design in collaboration with insurance market experts and product managers. Example project: Find ways to add uncertainty estimates to our flood parametric product by sampling additional data points spatially around the AOIs or with statistical resampling methods like bootstrapping.

- Deploy models into Floodbase’s monitoring pipeline on Google Cloud and own their continuous evaluation (provenance monitoring, reproducibility, etc.) in production.

Own our weather and hydrologic inputs: evaluate precipitation, reanalysis, discharge, and forecast products (satellite and radar rainfall, ERA5, GloFAS and EFAS, NWM, and ML weather models) and deploy transformations to compensate for sensor and dataset transitions in production without breaking the historical record.

- Evaluate recent advances in geospatial AI and incorporate them to improve the Floodbase product.

- Work closely with our team of machine learning engineers, research scientists, and software developers to translate rigorous science and research into robust products.

- Be a technical leader in the team, mentor engineers and scientists on the team and recruit other outstanding team members.

Minimum Qualifications:

- Bachelor’s in a quantitative discipline with 4+ years of industry experience or a graduate degree in a quantitative discipline with 1-2+ years of industry experience.

- Experience training deep learning models in PyTorch or a comparable framework, including at least one model taken from experiment into production.

- High coding proficiency in Python: can independently develop code and make and review contributions to a shared repository.

- Mastery of the fundamentals of scientific data analysis and statistics

Statistical judgment for product work: uncertainty quantification, resampling and calibration methods, and validation against sparse or noisy ground truth.

- Experience working in interdisciplinary teams solving complex problems that rarely have a clean, correct solution.

Preferred Qualifications:

The preferred qualifications describe the strongest candidate we can imagine, and are not meant to be a checklist. If you meet the minimum qualifications and some of the preferred ones, we encourage you to apply to this role.

- Familiarity with open-source geospatial ecosystems (rasterio/gdal, shapely, geopandas, xarray/rioxarray, Dask, Zarr).

- Demonstrated expertise in computer vision or multimodal sensor fusion from another domain, such as medical imaging, autonomous vehicles, robotics, or scientific instrumentation, where you have worked with noisy, partially observed, or irregularly sampled data.

- Familiarity with cloud ML infrastructure (GCP: GCS and Cloud Batch, or equivalents)

Hands-on experience with hydrometeorological data (precipitation products, reanalysis, modeled streamflow, or NWP and ML weather forecasts) and their characteristic errors

- Have a proven track record (open source projects, Kaggle, publications, etc) of developing new models for innovative applications.

- Working knowledge of recent GeoAI developments, including Earth-observation foundation models, pretrained embedding fields (Prithvi, Clay, AlphaEarth Foundations, TESSERA), and ML-based weather and hydrologic forecasting (GraphCast, GenCast, ECMWF AIFS, Aurora), with judgment about when they outperform task-specific models.

- Prior experience with satellite imagery (optical and Synthetic Aperture Radar) and/or geospatial modeling.

Why Join Floodbase: Join a company solving one of the most urgent problems of our time: flood risk in a changing climate - Work with cutting-edge geospatial data and AI technology - Competitive base salary + equity package - Influence the growth trajectory of a fast-scaling climate tech company

Compensation and Benefits:

Floodbase offers competitive salaries with regular reviews and adjustments based on performance and industry standards. The salary range for this role is commensurate with experience, with an expected range of $140,000 - $160,000. Additionally, this individual will receive equity compensation and a competitive benefits package.

Equal Opportunity Employer:

Floodbase is an equal opportunity employer and considers applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, age, or veteran status.