Applied Scientist – Forest Lidar & 3D ML

Applied Scientist – Forest Lidar & 3D ML WHAT IS SYLVERA ANYWAY? ‍👩‍👨🌳 Sylvera helps organisations make confident decisions in carbon and commodity markets. We independently assess the quality of carbon projects, track market pricing and supply signals, and provide geospatial and carbon intensity data — giving our clients the clarity they need to invest in real climate action. We're a team of 130+ scientists, engineers, and market experts based across London, New York, São Paulo, Singapore, and Tokyo. Our work spans market intelligence, geospatial analytics, multi-scale lidar research, and policy analysis. We partner with governments and policymakers to raise the bar for transparency and rigour across carbon markets. Backed by over $96 million from investors including Fidelity, Balderton Capital, Index Ventures, and Insight Partners, we're trusted by Fortune 500 companies, major financial institutions, and governments worldwide. If you care about climate, enjoy solving hard problems, and want to work with people who take the mission seriously — we'd love to hear from you. What will I be doing? ‍‍👩‍💻👨‍💻 We’re looking for a mission-driven Applied Scientist – forest lidar & 3D ML to join our Earth Analytics (EA) team. In this role, you will principally be tackling a fundamental research challenge within a high-impact commercial setting: automating the segmentation of individual trees from complex terrestrial laser scanning (TLS) point clouds. This work will directly impact and power Sylvera’s unique Biomass Atlas https://www.sylvera.com/products/forest-biomass-data product, which itself is becoming widely adopted for carbon accounting applications throughout carbon markets. To drive this breakthrough, you will have exclusive access to our world leading ground-truth, labeled dataset of 3D forest point clouds collected from forests around the world. Specific responsibilities will include: - Developing, training, and deploying 3D deep learning models (e.g., sparse convolutions, PointNet, TreeLearn or other similar architectures) to automate tree instance segmentation and Quantitative Structure Model (QSM) generation. - Translating experimental ML research into robust, reproducible code, working closely with our production engineering team to deploy models at scale. - Collaborating with our field teams and scientists to ensure model outputs align with biological reality and accurately capture complex forest structures. - Taking full ownership of an open applied research problem — scoping, prototyping, and iterating quickly to find solutions where no guaranteed blueprint exists. - Working in the lidar team to improve existing data products and methods to more efficiently undertake existing and ongoing manual segmentation and quality assurance. - And beyond segmentation, working with the wider EA team to solve other complex technical challenges involving point cloud and geospatial data, for example including retrieval of metrics characterising forest structure from high-density aerial lidar, forest carbon modeling, uncertainty quantification and automated tree species recognition. We’re looking for someone who: 🧠💚 - Has deep expertise in 3D point cloud processing and machine learning applied to 3D spatial data. - Is highly proficient in Python and the modern spatial/lidar software stack (e.g., PDAL, laspy, Open3D). - Brings domain knowledge in forest ecology - Cares deeply about the climate and ecosystems of the earth. - Is a self-starter who thrives in constantly evolving environments, ideally with early-stage experience. We’d like someone highly ambitious, motivated and eager to propel their career forward. We prioritise grit, positivity, and the willingness to get stuck in, and encourage you to apply even if your experience doesn't exactly match this job description. Benefits 💰 - Equity in a rapidly growing startup - Private Health Insurance and Life Assurance - Unlimited annual leave - and encour ... (truncated, view full listing at source)