AI Data Scientist
Locations: Pune, India
Buildings are getting smarter with connected technologies. With more connectivity, there is access to more data from sensors installed in buildings. Johnson Controls is leading the way in providing AI enabled enterprise solutions that contribute to optimized energy utilization, auto- generation of building insights and enable predictive maintenance for installed devices. Our Data Strategy & Intelligence team is looking for a Data Scientist to join our growing team. You will play a critical role in developing and deploying machine learning/Generative AI and time series analysis models in production.
The Role
To be successful in this role, the Data Scientist should have a deep knowledge of machine learning concepts, Large Language Models (LLM) including their training , optimization and deployment, time series models as well as experience in developing and deploying ML/Generative AI/ time series models in production.
What you will do
As an AI Data Scientist at Johnson Controls, you will help develop and maintain the AI algorithms and capabilities within our digital products. These applications will use data from commercial buildings, apply machine learning, GenAI or other advanced algorithms to provide value in the following ways:
- Optimize building energy consumption, occupancy, reduce CO2 emissions, enhance users’ comfort, etc.
- Generate actionable insights to improve building operations
- Translate data into direct recommendations for various stakeholders
Your efforts will ensure that our AI solutions deliver robust and repeatable outcomes through well-designed algorithms and well-written software.
To be successful in this role, the AI Data Scientist should be comfortable applying machine-learning concepts to practical applications while handling the inherent challenges of real-world datasets.
How you will do it
- Contribute as a member of the AI team with assigned tasks
- Collaborate with product managers to design new AI capabilities
- Explore and analyze available datasets for potential applications
- Write Python code to develop ML/Generative AI/time series prediction solutions that address complex business requirements
- Research and implement state-of-the-art techniques in Generative AI solutions
- Pre-train and finetune ML over CPU/GPU clusters while optimizing for trade-offs
- Follow code-quality standards and best practices in software development
- Develop and maintain test cases to validate algorithm correctness
- Assess failures to identify causes and plan fixes for bugs
- Communicate key results to stakeholders
- Leverage JIRA to plan work and track issues
What we look for
- Bachelor's / Master’s degree in Computer Science, Statistics, Mathematics, or related field.
- 5+ years of experience of developing and deploying ML Models with a proven record of delivering production ready ML models.
- Proficiency with Python and standard ML libraries, e.g., PyTorch, Tensorflow, Keras, NumPy, Pandas, scikit-learn, Matplotlib, Transformers.
- Strong understanding of ML algorithms and techniques, e.g., Regression, Classification, Clustering, Deep Learning, NLP / Transformer models, LLMs and Time Series prediction models.
- Experience in developing SOA LLM frameworks and models (Azure OpenAI, Meta Llama, etc), advanced prompt engineering techniques, LLMs fine-tuning/training.
- Experience in working with cloud (AWS / GCP / Azure) based ML/GenAI model development / deployment.
- Excellent verbal and written communication skills.
Preferred
- Prior Domain experience in smart buildings and building operations optimization
- Experience in working with Microsoft Azure Cloud.
