Internship

Logo for Pahk Solutions Inc.

Machine Learning Engineer

Pahk Solutions Inc.


  Publié: May 13, 2022

  Emplacement: Toronto (with a remote working option), Ontario

  Échelle salariale: $40,000 - $50,000

  Heures par semaine: 30

  Date de début: June 15, 2022

  Date de fin: February 28, 2023

Description de l'emploi

Greenhouse gas (GHG) emissions from space heating and cooling of residential buildings account for one of the largest sources of household environmental footprint. In 2020, 52% of the residential heating was powered by natural gas in Canada, which resulted in generation of ~43 mega tonnes of CO2.

Retrofitting the existing buildings is one the most prominent ways of improving building efficiencies and meeting the climate-positive targets. It has been shown that retrofits in Canada can result in a median 16% reduction in total buildings energy consumption. Conventionally, determining a building envelope’s energy efficiency and identifying viable retrofit candidates relied on on-site performance appraisals and complex simulations. However, prevalence of internet of things (IoT) sensors, access to big data as well as advancement in the field of data analytics and machine learning have resulted in innovative alternatives to conventional methods.

In this work, we aim to use IoT device data and machine learning techniques to identify the heat transfer behavior of building envelopes and determine viable retrofit candidates. For determining retrofit candidates, we first observe the building envelopes thermal behavior based on the data collected from IoT devices and external sources (e.g., building modeling data, outdoor temperatures). Further, we use machine learning and statistical models to classify and benchmark the buildings thermal behavior and their characteristics. This analysis enables us to identify highest-impact retrofit opportunities.

Devoirs et responsabilités

- Developing API integrations to pull data from IoT devices
- Gathering and standardizing ancillary datapoints from open-source datasets including building designs, building codes, etc.
- Developing first-order exponential decay thermal model of buildings based on data from IoT devices
- Researching, developing and implementing machine learning algorithms to identify retrofit candidates by classifying buildings thermal behavior
- Performing statistical analysis to improve models
- Selecting best data visualization methods and creating reports for management other team members
- Working closely with other members of the technical team for implementation of machine learning models

Connaissances et compétences

- Advanced math and statistics skills, surrounding subjects such as linear algebra, calculus and Bayesian statistics
- Bachelor’s degree in computer science, math, statistics or an engineering/science degree with related experience
- Solid understanding of thermodynamic principles
- Strong analytical, problem-solving and teamwork skills
- Experience in data science
- Experience with coding and programming languages, including Python, Java, C++, C, R and JavaScript
- Experience in working with ML frameworks
- Experience working with ML libraries and packages
- Understand data structures, data modeling and software architecture

Éducation requise

Bachelor's Degree in Engineering, Applied Science or related fields

  Comment s'inscrire

S'il vous plaît envoyer un courriel tina@pahk.io pour postuler au poste ci-dessus.

Pour postuler, envoyez votre CV à tina@pahk.io.

Que disent les gens à propos de Lancement de carrière?

Nos stagiaires et nos employeurs partagent leur expérience de première main.

Voir tous les témoignages
Quote marker image

J’ai toujours voulu travailler dans le domaine des soins de santé, et ce programme et ce stage m’ont permis de mettre un pied dans le métier.

– Madison Smith, étudiante


Quelques-uns de nos employeurs fantastiques que nous avons aidés

Notre financement leur permet de contribuer à remédier aux pénuries de main-d’œuvre, à diversifier leur personnel et à renforcer leur secteur.

Commencer