About KaleidEO
KaleidEO, a subsidiary of SatSure, is investing in launching its own fleet of earth observation satellites. We are seeking an Image Processing wizard to own and nurture the logical pipeline that delivers the best geospatial images and products. SatSure is a deep tech, decision Intelligence company that works primarily at the nexus of agriculture, infrastructure, and climate action creating an impact for the other millions, focusing on the developing world. We want to make insights from earth observation data accessible to all. The synergy of KaleidEO and SatSure aims to bring a new dimension to the earth observation industry by being the only full-stack company from India, to have satellites in space to deliver insights on the ground. If you are interested in working in a space tech analytics domain driven by cutting-edge technology on the edge as well as on the ground for solving complex data problems using Multi-Sensor Earth Observation data you will have the freedom to work on innovative ideas and be creative with no hierarchies, KaleidEO us the place for you.
Role
- Implementing deep-learning models and/or frameworks that can be used in scene/semantic/panoptic segmentation, object classification, and synthetic data generation using High-Resolution Earth Observation Data.
- Design and implement performance evaluation frameworks for incremental and step-function improvements.
- Exploring and developing new methods for improving detection/classification performance.
- Building model monitoring and active learning pipelines to address the target and data shifts happening in the domain.
- Building models to solve open research problems on complex image preprocessing, computer vision, object identification, segmentation, change monitoring, multi-source data fusion, etc. from High-Resolution Earth Observation data sources.
- Define and implement performance metrics for the research problems and be very methodical / organized with experiment management and performance evaluations.
Requirements
Experience:
- 3-5 years of experience in machine learning-based data product research and development in addition to a relevant educational background in the field of Computer Science, Mathematics, Engineering, or Machine Learning.
- Experience with tools/frameworks from theoretical and practical aspects of computer science and ML/DL. This includes calculus, probability, statistics, learning theory, etc.
- Prior experience and core foundational knowledge on working with statistical modeling techniques and deep learning fundamentals, such as CNN, RNN, Transformers, GANs, etc. including sequence networks eg LSTMs
- Proficiency in Python or C++, with hands-on experience in using ML frameworks such as Tensorflow/Pytorch, Pytorch Lightning, MxNet, etc
- Ability to design, develop, and optimize for machine learning/ deep learning models.
- Knowledge of the application of ML for earth observation data is good to have.
- Appetite for research and implementing innovative solutions for complex problem
Competencies:
- A proven ability to learn new tools and technologies quickly.
- Ability to develop novel solutions rather than look for readily available options.
- An ability to translate complex topics and tools into easy-to-understand concepts in order to explain the impact of developed models to stakeholders and peers.
- Excellent debugging and critical thinking skills.
- Excellent analytical and problem-solving skills.
- Ability to work in a fast-paced, team-based environment.
Educational Qualifications:
- Master's degree or PhD in Engineering, Statistics, Mathematics or Computer Science, Machine Learning / Deep Learning domains
Benefits
Why Us
- Opportunity to work on a unique and futuristic technology setup
- Flat organizational structure and accessibility
- Best-in-class leave policy
- Additional allowances for learning, skill development, broadband, medical insurance cover, etc.
