Machine Learning Engineering Intern

Job Description

Wikilmo is an early stage Climate Informatics startup working to build solutions that help augment climate resilience in remote farming communities through the use of data-based predictive actionable insights delivered to the last mile. Our ongoing projects involve developing hazard monitoring solutions, estimating rainfall patterns, harmonising agricultural data, identifying and predicting possible pest outbreaks, all focussed towards delivering insights to remote locations that only have limited infrastructure and services. We are also developing an app that integrates all of these solutions into a single light-weight, intuitive and fully online-offline experience.

This is a remote internship position for 6-8 weeks with the expectation of at least 15 hours of work per week. There is a token stipend paid to the interns.

Qualifications

Requirements
We are hiring for multiple positions and expect candidates to be proficient in at least some of the following tasks. Proficiency is demonstrated through previous internships, coursework, projects and participation in hackathons & competitions.
- The intern will research and model aspects of the environment utilizing satellite imagery, LiDAR, Radar and other datasets.
- Explore computer vision algorithms that extract insights from aerial imagery and geospatial data
- Process precipitation estimation and NDVI products derived from Satellite Imagery/Radar data sources
- Implement existing research and improve models for better quantitative estimation of weather conditions and pest outbreaks

Preferred Skills
- Experience with implementing and optimising Machine Learning models from research to deployment stage
- Familiar with metrics involved in ML, algorithm selection, and cross validation
- Experience with satellite imagery/weather/time series datasets using in climate or agri informatics
- Experience with version control, documentation and software development practices
- Willing to engage in rigorous code reviews and give/receive friendly, constructive criticism for the sake of creating high-quality software