Carbon Emissions Forecasting - Master Student

We are a CleanTech startup building the next generation of digitalized and environmentally aware electrical grids. To do this, we believe artificial intelligence has the potential to help society solve immense global challenges such as the transition to a greener future.

Emissium invites talented students at both bachelor's and master's levels to embark on an exciting journey of collaboration. As part of our team, you'll play a pivotal role in shaping our next-generation machine learning algorithms, and influencing the decisions of major energy consumers! This project spans the entire machine learning lifecycle, from data cleaning and preprocessing to model training, testing, and potential deployment to production.

Key project components:

In a world where robust machine learning (ML) models drive critical decisions for large energy consumers, your role becomes essential. The project focuses on gaining insights into the electricity and environmental impact data of the Emissium network. You will work on data preprocessing, investigating techniques for feature selection, such as correlation analysis and recursive elimination, or leveraging domain knowledge to guide the selection process. You will develop strategies to handle missing data using conventional or more advanced ML-based imputation techniques. You will train diverse ML models for multivariate time-series forecasting, evaluate model robustness and their performance metrics (accuracy, F1 score, ROC-AUC), discussing the importance of selecting metrics based on the problem context.

If your models demonstrate good performance, you will be able to deploy them into production! In this case, you will explore deployment options (cloud services or on-premises servers), work on model serialization (e.g., pickle for Python, ONNX, TensorFlow SavedModel), create APIs (Flask, FastAPI, or Django), end ensure scalability (handle the expected load and scale horizontally if necessary, use load balancers for distributing incoming requests). You will learn how to develop comprehensive documentation (API, usage instructions, input/output formats, and potential issues) and considerations for monitoring (model performance, input data distribution, and system health) and logging (record predictions, errors, and other relevant information).

Why choose Emissium:

Real-world impact: Contribute to critical decision-making in energy consumption.
Professional growth: Immerse yourself in cutting-edge technologies and the latest developments in AI.
Career opportunities: Exceptional performers may have the chance to join the Emissium team.

What we seek:

A collaborative and passionate individual with a keen interest in artificial intelligence.
Thoughtfulness about AI's societal impact, especially in the context of the energy transition.
Previous knowledge of deep learning frameworks (PyTorch or TensorFlow), Python, and ML libraries (e.g., scikit-learn and pandas) are welcomed.
Interest in visualizing and manipulating big datasets.
Basic knowledge of English to thrive in our international team.

Emissium is committed to supporting you throughout the project and welcomes long-term engagement with students delivering exceptional results. Apply now to be part of a dynamic team, shape the future of energy, and open doors to exciting career possibilities!