Principal Data Scientist - Mumbai, Maharashtra, India - TVS Motor

TVS Motor

Verified Company

Mumbai, Maharashtra, India

1 week ago

Posted by:

Deepika Kaur

beBee Recuiter

Description

Group Company:
TVS Motor Company

Designation:
Principal Data Scientist (PDS_D&AI - Common Engineering, DPM - CE)

Office Location:
Electronic City (Territory), Karnataka Koramangala (Territory)

Position description:

Need to deliver solutions to production through the software development lifecycle

Provide leadership to establish world-class ML lifecycle management processes.

Write production ready code and deploy real time ML models; expose ML outputs through APIs

Partner with data/ML engineers and vendor partners for input data pipes development and ML models automation

Good to have hands-on experience with open source tools such as Kaldi, Py torch-Kaldi.

Primary Responsibilities:

Build ML/AI models leveraging a strong understanding of Machine Learning principles including standard algorithms for Regression and Classification, Deep Learning constructs
Build recommender systems around agent Collections prioritization, product recommendations, and personalization engines.
Build sophisticated pricing engines based on AI powered product evaluation tools.
Build crosssell/upsell recommender systems using deep learning frameworks, and with sparse data.
Able to understand the determinants of success for a Machine Learning system, Model accuracy and efficiency, Data requirements, Training and Test constructs, CI/CD for ML systems.
Able to build standard ML systems using available ML components provided by AWS, Azure and GCP.
Cloud & Big Data:
1. Work on Cloud Infrastructure (Azure, AWS) to provision data to ML models, build ML systems on deploy them at scale.
2. Build ML systems using Spark libraries such as ML Lib, Spark SQL and be able to deploy them on clusters/machines, both on Cloud and onPrem.
3. Put in place systems to continuously monitor, evaluate and retrain ML models in production.
1. Design and implement the machine learning lifecycle at scale, from building the data infrastructure to train/test Machine learning models to their production environments.
2. Management of production ML workflows ensuring automated CI/CD capabilities are built into the workflow.
3. Design and Implement alerts/dashboards to ensure continuous monitoring of production models' effectiveness (accuracy, latency, performance etc.,)