Who we are:
Pachama is a mission-driven company looking to restore nature to help address
climate change. Pachama brings the latest technology in remote sensing and AI
to the world of forest carbon in order to enable forest conservation and
restoration to scale. Pachama’s core technology harnesses satellite imaging
with artificial intelligence to measure carbon captured in forests. Through
the Pachama marketplace, responsible companies and individuals can connect
with carbon credits from projects that are protecting and restoring forests
worldwide.
We are backed by mission-aligned investors including Breakthrough Energy
Ventures, Amazon Climate Fund, Chris Sacca, Saltwater Ventures, and Paul
Graham.
Recent press:
Pachama is #1 most innovative AI company
Jeff Bezos' Last Shareholder Update
Pachama to monitor and manage Mercado Libre forest projects
We are looking for a Senior Machine Learning Engineer to lead the development
of cutting-edge systems for our mission to restore nature to help solve
climate change. As a leader on the Science team, you will build, scale and
deploy AI and remote sensing technology to create products to identify and
originate high-quality forest carbon projects. A typical day includes
implementing new machine learning models with remote sensing and other
geospatial data, designing experiments to validate their performance, pair
coding with other engineers, and discussing results and experiment plans with
scientists. The quality of model outputs directly impacts the quality of
forest carbon projects. Model validation and uncertainty quantification are
core values for our team. Tracking and leveraging innovations described in
science papers and from commercial applications is a critical component of
this role.
We're looking for engineers who find joy in the craft of building but live to
see the end-to-end impact and want to rally engineers around them. Engineers
who push forward initiatives by asking great questions, cutting through
ambiguity, and organizing to win. Engineers who are relentlessly detail-
oriented and methodical in their approach to understanding trade-offs place
the highest emphasis on building quickly.
Location:
This role is remote within North American time zones only.
What You Will Help Us With:
* Training machine learning models to estimate key forest structure parameters essential to quantify ecosystem carbon storage and evaluate the climate benefit of forest carbon projects. Work with the Product team to align product value with scientific and technical complexity.
* Advocating for and mentoring on best practices applied to our AI and data science work. Mentoring teammates to raise the bar across the Science and Engineering teams to enable step-level efficiency, accuracy, and reliability increases[.](http://reliability.Designing)
* Designing statistical frameworks and experiments to assess the accuracy and uncertainty of these models on real-world [data.](http://data.Optimizing)
* Optimizing these models to run efficiently on large amounts of geospatial and remote sensing [data.](http://data.Helping)
* Helping construct tools enabling research and operations to produce high-quality performance metrics for forest carbon projects.
* Clearly communicating the impact and learnings from our deep technical work cross-functionally so organizationally, we understand how AI and remote sensing can help us find and design better projects.
Experience & Skills We’re Looking For:
* Machine learning and statistics fundamentals with an ability to apply these skills to domains like forest science and remote sensing.
* Expertise deploying deep learning models at scale using distributed computing.
* Strong software engineering practices and a background in Python programming, debugging/profiling, and version control. Some examples of tools in our tech stack include Kubernetes, Dask, Flyte. Open source geospatial tools that are also part of our tech stack include Rasterio, Geopandas, and Xarray.
* Expertise working in cluster environments and an understanding of the related distributed systems concepts (CPU/GPU interactions/transfers, latency/throughput bottlenecks, pipelining/multiprocessing, etc).
* Experience working with and processing LiDAR datasets for vegetation structure and analysis.
* Experience working with terrestrial ecosystem geospatial data such as land cover, ecosystem classifications, and biophysical data.
* Experience working with Landsat, Sentinel 1 and 2, and high-resolution imagery (e.g., NAIP, Planet, etc.). Knowledge of harmonizing imagery across optical sensors
* Ability to find and synthesize related academic literature to apply these learnings to model and experiment design.
* Comfort with fast-paced execution and rapid iteration startup environment. Excited by product impact.