Voltaware is a UK-based energy monitoring start-up, making a breakthrough in the IoT market by proposing a technology combining:
• behavioural monitoring for elderly care,
• breakdown of the electricity consumption for individuals,
• preventive maintenance for businesses.
Based in the UK but with collaborators across 4 continents, Voltaware a welcoming multicultural workspace. Over the last 6 years, Voltaware’s team incorporated a dozen of interns and junior profiles and helped them develop brilliant careers in data science.
Work language
100% English
Development and support
With a growing data team of already 10 people, and with the support of experienced python developers, infrastructure engineers and devops, Voltaware is a place where pertinent ideas are meant to be deployed to production.
As a startup, Voltaware used to train people by giving them responsibilities. This role fits particularly to a students who wants to be confronted to challenges of commercial applications of data science.
The Project: Clustering of Energy Consumption Data
Clustering is a vast subfield of unsupervised learning with applications in Biology, Social sciences, Digital Marketing …
In the field of Energy data, clustering is made more complex by the nature of the data: a timeseries representing a discrete state machine. In addition, the literature is limited to covering at most a few hundreds of homes.
For this project, we propose to experiment clustering technics on thousands of household timeseries consumptions. The goal will be to use the clustering to answer the following questions:
• How many clusters do we need to describe/summarise a population of energy consumers?
• Can we fill the nan’s contained in an energy consumption time series using some clustering technic?
• How do energy consumption clusters compare with household profiles (number of people, number of electrical appliances…)?
• Any other relevant question raised by the intern!
In addition to the algorithm development and experiment, the intern will be expected to develop the tools to evaluate and visualize their work.
Internship conditions
• Location: The internship will be completed remotely with the possibility of weekly meetings with the Head of Data Science in our Madrid office.
• Support: the intern will be directly supervised by Voltaware’s Head of Data Science and will have all the necessary support from the rest of the team. For the duration of the internship, the intern will be considered as a full member of the team and will have access to Voltaware development platform.
• Recruitment opportunities: in case of success, the intern will be offered a permanent position inside the team.
Requirements
• To be enrolled in a master program centred around data science, computer science or applied math,
• To have at least an intermediate level in python and python data science libraries (numpy, pandas and sci-kit learn),
• Excellent written and oral communication in English,
• Strong motivation to work in a green start-up.
Bonus skills and experiences
• Education or experience in electrical engineering or electronics,
• Experience with time series analysis or with spectral analysis,
• Data engineering experiences and skills are always welcome.
Selection steps
- 1h interview with Voltaware’s Data Lead (company and candidate introduction followed by a few data related questions)
- Data science test: a sample of data is given to the candidate along with a couple of questions. After a few days, the candidate presents in 30 minutes their code and analysis of the data.
Benefits
• Compensation of 1000 euros per month based on 20 hours of work per week.
How to apply:
• by sending cv and cover letter to: ds-recruitment@voltaware.com
