Introduction
We are hiring a Data Science Intern.
You have completed your bachelor's degree in engineering, computer science, mathematics, or a related quantitative field
You’ve demonstrable experience working with time series forecasting techniques and libraries (pandas, scikit-learn, statsmodels, TensorFlow/PyTorch, MLflow)
Prior experience working with energy data, smart meter data, or IoT sensor data is highly preferred, but not a must-have
Strong programming skills in Python with experience in data manipulation, statistical analysis, and machine learning
Someone who thrives on experimentation and can work independently while collaborating effectively with technical teams
Available to start on short notice and eager to work in a fast-paced, innovative environment focused on solving one of the most pressing problems preventing climate change
Where you'll work and how you'll contribute:
Support the development of a new product line that we're launching soon, where the forecasting models you build will be a core component
Solve real-world problems in demand-side management and help the energy transition by preventing the need for rapid grid capacity expansion through accurate energy demand predictions
Work collaboratively with a cross-functional team comprising software engineers, product managers, data scientists, and building physics and energy experts
Develop and experiment with multiple time series forecasting models to predict building energy demand patterns
Analyze smart meter data and building performance metrics to identify key features and patterns that influence energy consumption
Implement and compare various forecasting approaches (ARIMA, XGBoost, Prophet, etc.) to determine the optimal modeling approach
Conduct data preprocessing, feature engineering, and model validation to ensure robust and accurate predictions
Create visualizations and reports to communicate model performance, insights, and recommendations to stakeholders
Spectral is a smart energy systems and platforms integrator driven by purpose. Presently, the grid's capacity for renewable energy is limited (10-20%) because of the fluctuating supply from renewable sources, to maintain stability. A successful clean energy transition hinges on innovative approaches to energy management and new business models. Spectral develops technological solutions for energy storage, demand flexibility, and local renewable energy to maximize their potential and build better smart grids.
As Spectralites, we are dedicated to accelerating the world's transition to a sustainable energy future. Achieving such a large-scale system change will require a monumental effort from everyone, and it will undoubtedly present significant hurdles. Rapid electrification across sectors and renewable energy expansion are creating grid congestion and complex balancing challenges. Our team is addressing the fascinating but difficult issue of energy inefficiency in modern structures, even with improved standards, focusing on poor climate system management and insufficient data transparency. By developing and deploying advanced technology platforms, we're driving faster change in real estate and energy, resulting in well-connected, more sustainable, and high-value assets.