Join BeZero Carbon as they are looking for a Geospatial Data Scientist
Overview
No salary declared 😔
EC1Y 8QE
Expires at anytime
About us:
BeZero Carbon is a global ratings agency for the Voluntary Carbon Market. Our ratings allow all market participants to price and manage risk. BeZero’s ratings and research tools support buyers, intermediaries, investors, and carbon project developers.
Founded in April 2020, our 120+-strong team combines climatic and earth sciences, sell-side financial research, earth observation, machine learning, data and technology, engineering, and public policy expertise. We work from five continents.
Background on the role:
BeZero is looking for a mid- or senior-level data scientist to join our geospatial machine learning team. The team is responsible for building and machine learning products for our client-facing platform and internal teams. The team sits at the heart of the business, working across our ratings, product, and technology teams and we plan to grow the team with several new members.
As a data team, we have a bias towards shipping products, staying close to our internal and external customers, and owning our infrastructure and deployments end-to-end. This is a team that follows software engineering best practices closely. Our data stack looks as follows:
-AWS as our cloud infrastructure provider.
-Snowflake as our central data warehouse for tabular data. AWS S3 is used for our raster data, and we use PostGIS for querying geospatial vector data.
-dbt for building SQL-style data models and taking care of the ‘T’ in ELT.
-Python jobs for non-SQL data transformations and modelling. For geospatial processing we heavily use packages like gdal, rasterio, xarray, geopandas, and for modelling we use anything relevant in the Python scientific computing ecosystem (e.g., scikit-learn, PyTorch).
-AWS Sagemaker for model development and deployment.
-Prefect as our workflow orchestration manager, with our jobs executed on AWS ECS.
-Metabase as a dashboarding solution for end-users.
-GitHub Actions for CI / CD.
You’ll be responsible for building geospatial algorithms that directly affect the way our ratings team analyses the quality of carbon offset projects. You’ll therefore work closely with researchers in our Geospatial Research team and ratings scientists in our Ratings team. We process large-scale satellite imagery data sets (think about any of the public NASA and ESA missions) of different types (optical imagery, radar, SAR) for most of our machine learning use-cases, but also leverage raster and vector data from partnerships we have with data vendors. You'll most likely be someone that has experience building machine learning (deep learning) algorithms for large-scale imagery processing.
You’re welcome to work in our London-based office, but we very much welcome applications remotely if you can work GMT +/- 3 hours[1] .
Responsibilities:
You will be an individual contributor in our geospatial machine learning team, focused on developing and maintaining geospatial machine learning products to be deployed on our carbon markets platform or used internally by our ratings team.
You will research and implement statistical and machine learning (deep learning) approaches to solve business challenges relevant to carbon markets.
You will work in a collaborative, production-facing codebase that has close coupling with engineering systems.
You will work with our internal research and ratings teams to integrate the outputs of (analytical) data pipelines into BeZero’s business processes.
You’ll be our ideal candidate if:
You care deeply about the climate and carbon markets and are excited by solutions for decarbonising our economy.
You are a highly collaborative individual who wants to solve problems that drive business value.
You have 3+ years of experience building statistical or machine learning algorithms in a commercial setting and deploying these in production.
You have experience in building and deploying deep learning models for computer vision use-cases.
You have experience dealing with a variety of geospatial data formats (e.g., netCDF, (cloud-optimised) geotiff, geoJSONs, zarr) and geospatial SQL and Python packages (e.g., PostGIS, xarray, rasterio, shapely, gdal).
You can write clean, maintainable, scalable, and robust code in Python and SQL, and are familiar with collaborative coding best practices (e.g., for Python PEP8 code style, unit testing, continuous integration tools such as flake8, black, isort, etc).
You have experience with a workflow orchestration tool (e.g., Prefect, Dagster, Airflow, Luigi), a cloud platform (we use AWS but any cloud platform will do) and the Python scientific computing stack (NumPy, SciPy, matplotlib, pandas, etc).
Bonus points (but we’d still like to hear from you if you don’t have experience in any of these)
You have experience with machine learning platforms (like AWS Sagemaker, MLFlow, weights & biases) to design, test, serve and monitor geospatial ML pipelines.
You have deep experience in analysing specific satellite imagery data types like SAR, LiDAR, or RADAR or another remote sensing domain.
Research has shown that women are less likely than men to apply for a role if they don’t have experience in 100% of the requirements outlined in a job description. Please know that even if you don’t have experience in all the areas above but think you could do a great job and are excited about shaping company culture, finding great people, and building great teams, we’d love to hear from you!
What we’ll offer:
- Competitive salary and opportunity for equity in a rapidly growing VC-backed start-up through share options
- Ability to learn and develop alongside a range of sector specialists from the scientific, economic and business community
- Opportunity to work in a cross-cutting role, interacting with lots of different parts of the business
- Growth opportunities that come from working at a fast-paced VC-backed technology business
- Opportunity to work remote or in our Central London office space (Old Street) with flexibility to work from home + some flexibility over working location during the summer
- Regular social events
- 25 days leave (with additional time off between Christmas and New Year, and for your birthday)
- Private medical insurance, dental, critical illness cover, income protection, life assurance, medical cash plan, cycle to work scheme, and a health and wellness cash allowance
We value diversity at BeZero Carbon. We need a team that brings different perspectives and backgrounds together to build the tools needed to make the voluntary carbon market transparent. We’re therefore committed to not discriminate based on race, religion, colour, national origin, sex, sexual orientation, gender identity, marital status, veteran status, age, or disability.