Getir, established in 2015, is a leading ultrafast grocery delivery company with a multinational presence. Utilizing Amazon SageMaker and AWS Batch, they reduced model training time by 90% and improved operational efficiency. Their data science team developed a product category prediction pipeline with an 80% accuracy rate, aiding commercial teams in inventory management and competitive benchmarking. The process included refining multilingual BERT models and employing Amazon S3 for model storage.
“`html
Getir’s AI Revolution in Ultrafast Grocery Delivery
Getir has become a leading name in ultrafast grocery delivery since its inception in 2015. Expanding from Turkey to global markets, including the UK, the Netherlands, Germany, and the USA, Getir has established a strong multinational presence with a diverse range of services.
Streamlining Operations with Amazon SageMaker and AWS Batch
The team at Getir has developed an end-to-end product category prediction pipeline that has significantly reduced model training duration by 90%. This innovation aids the commercial teams by providing data-driven insights for inventory management and customer interactions, enhancing the company’s competitive edge.
Challenges and Solutions in Machine Learning
Getir’s data science team faced the challenge of understanding their product assortment in comparison to competitors. They overcame this by:
– Utilizing AWS tools for machine learning and predictive analytics.
– Fine-tuning BERT models for seven different languages using AWS Batch GPU jobs.
– Efficiently storing models using Amazon S3 for easy access and deployment.
Operational Efficiency and Future Exploration
The successful integration of SageMaker and AWS Batch has led to:
– An 80% prediction accuracy across all category levels.
– Reduced model training durations, enhancing operational agility.
– Plans to explore SageMaker multi-model endpoints (MMEs) for even more efficient model management.
Getting Started with Your Own ML Pipelines
For those interested in building their own ML pipelines, resources for Amazon SageMaker and AWS Batch are available to help you get started.
About the Authors
The post is co-authored by experts from Getir and AWS, who bring a wealth of knowledge and experience in data science, engineering, and machine learning.
Evolve Your Company with AI
If you’re looking to stay competitive and harness the power of AI, consider how Getir achieved a 90% reduction in model training durations with Amazon SageMaker and AWS Batch. Discover AI solutions that can transform your business operations and customer engagement.
– Identify Automation Opportunities: Pinpoint customer interaction points for AI enhancement.
– Define KPIs: Set measurable goals for your AI initiatives.
– Select an AI Solution: Choose tools that meet your specific needs and allow for customization.
– Implement Gradually: Start small, analyze data, and expand AI use wisely.
For expert advice on AI KPI management, contact us at hello@itinai.com. Stay updated with AI insights on our Telegram t.me/itinainews or Twitter @itinaicom.
Spotlight on a Practical AI Solution
Explore the AI Sales Bot at itinai.com/aisalesbot, designed to automate customer engagement around the clock and manage interactions throughout the customer journey. Visit itinai.com for more AI solutions.
“`
