Data Scientist in Bioacoustics
Help companies understand their impact on biodiversity by using acoustic analysis to measure and characterize the ecological health of a location
About the job
Green Praxis helps companies with large land footprints manage their land in a sustainable and economically profitable way. We offer Nature Based Solutions to solve a variety of environmental challenges and help customers understand the tradeoffs when planning interventions between economic and environmental indicators such as Biodiversity, Carbon Sequestration & Risk. The company consists of scientific experts in plants, animals and soil complemented by a technology team that helps them perform analysis at scale and uncover insights.
​
We are investing in a research program to further develop our Bioacoustics capability. Bioacoustics is the measurement of species richness and abundance based on analysis of the soundscape of a location. Preserving biodiversity is critical to our planet's health and a critical component of reducing environmental impact is the ability to measure that impact. Bioacoustics is an extremely promising field that has the potential to tell us a great deal about the natural life in a landscape using affordable technology.
We are looking for an Data Scientist in Bioacoustics to support these efforts. In this role you will develop the sound analysis backbone to support this work, in particular by:
-
Creating a data processing pipeline for sound processing across a large number of audio files (we record 24 hours/day for multiple days across multiple sites with multiple high quality microphones - it's a lot of data!)
-
Implementing signal processing techniques to improve the quality of the data for tasks such as tagging or removing wind noise, rain noise and human disturbance
-
Implementing, analyzing and optimizing the creation of acoustic indices that measure different characteristics of the sound signal, and research their relationship to animal vocalizations
-
Develop techniques for the recognition of keystone species
-
Implementing machine learning techniques to cluster, label and compare different soundscapes and build out a catalog of sound environments that can be used for future classification
In addition to this core work, you will also be asked to contribute to other parts of our IT portfolio. We have a mobile and web app running on a cloud infrastructure and multiple models related to vegetation. We are a small team so we all pitch in when and where needed. Our computing environment is a mix of Python, Dart and NodeJS running on IBM, Amazon and Google infrastructure.
​
What we are looking for in you:
-
PhD or Masters degree with concentration in data science or a related discipline
-
2+ years leveraging machine learning in real world applications
-
Strong interest in ecology and willingness to invest in building domain knowledge
-
Demonstrated experience of building data pipelines
-
Demonstrated experience of building models and leveraging machine learning techniques
-
Demonstrated experience of audio or image signal processing
-
Good understanding of statistics
-
Good understanding of core IT principles
-
Ability to work with substantial autonomy
-
Good communicator and collaborator
-
The company is French but we have a mix of nationalities so our working language is English. You will need to be proficient in written and spoken English.
If this sounds like you, please get in touch!