Client:
Altigreen
Introduction:
Altigreen, based in Bengaluru, specializes in manufacturing electric 3-wheelers tailored for last mile transportation in India. They design, engineer, and produce technologically advanced vehicles that cater specifically to local road conditions and driving habits. Altigreen’s electric cargo vehicles mark a significant advancement in India’s electric vehicle sector.
Business Need:
The customer faced significant challenges in their ML model development, particularly in terms of infrastructure management, scalability, model development efficiency, and MLOps and deployment. Traditional methods, along with the utilization of EC2 and ECR, posed difficulties in effectively addressing these challenges.
Solution Approach:
Rapyder conducted a series of strategic meetings with the Alitgreen team to thoroughly understand their challenges. Leveraging this insight, Rapyder implemented a tailored solution using Amazon SageMaker, which streamlined Alitgreen’s workflows and accelerated innovation in their electric vehicle development.
Implementation:
- Implemented a scalable model building and training, expediting development with built-in algorithms, enhancing efficiency through automatic model tuning, and providing seamless deployment and retraining functionalities.
- Effectively addressed challenges associated with ML Model development to accelerate innovation in electric vehicles and last-mile transportation.
- Amazon S3 bucket and AWS Lambda functions are utilized to trigger batch transform jobs for a batch prediction, allowing for efficient prediction of the data.
- To keep stackholders informed, email notifications are integrated into the workflow. These notifications provide critical updates on the success, failure, and execution of various steps in the process.
- ML workflow solution optimizes the model development and deployment process by leveraging Sagemaker pipelines, automated triggers, model approval stages, and email notifications.
Industry:
Automobile
Offering:
MLOps
AWS Services:
Amazon S3, Amazon SageMaker, AWS CodePipeline,
AWS Lambda, AWS CodeCommit, AWS CodeBuild,
Amazon EventBridge,
AWS CloudFormation
Reaping Rewards:
- The customer experienced reduced model preparation and deployment times, even when handling large datasets.
- Faster turnaround and consistent model performance led to improved service quality, enhancing customer satisfaction and retention.
- The adoption of cutting-edge technology and the ability to rapidly deploy new models enabled the customer to stay ahead of the competition, attracting a broader customer base.
- Reliable workflows facilitated smooth model deployments, driving continuous business growth and fostering innovation.
- Task automation minimized technical overhead, saving time and resources, allowing the customer to focus on business expansion.
- Consistent and reliable workflows ensured seamless model deployments, contributing to sustained business growth and innovation.
