Agenda

• Breakout tracks [1:30 PM - 5:00 PM IST]

  • Track 1: Generative AI on AWS
  • Track 2: AI & ML stories
  • Track 3: AI & ML for business applications
  • Track 4: AI and ML solutions

  • Track 1: Generative AI on AWS

  • Build your generative AI application with Amazon Bedrock
    

    We are at an exciting inflection point in the widespread adoption of machine learning (ML); with the growth of generative AI applications. In this session, learn how to build your generative AI application with key services including Amazon Bedrock. Get tips on getting started quickly. Learn from reference architectures for common use cases built with AWS AI & ML tools and services, such as self-service customer support, text analysis, report generation, post-call analysis, and forecasting trends.
    

    Speaker: Harshal Pimpalkhute, Senior Manager Product, Generative AI, AWS
    

    Time: 1:30 PM  - 2:00 PM

    ---

    Innovate on enterprise data with Amazon Q to improve employee productivity
    

    Bringing the power of generative AI to your employees in a secure and fast way can be challenging. Amazon Q is a new type of assistant powered by generative AI that is designed for work and can be tailored for your business. In this session, we'll explore how Amazon Q can provide secure, quick access to answers from the content in your enterprise and help improve productivity for your employees. Learn how you can setup and use Amazon Q to get contextual answers from connected data sources, summarize documents, generate content , and automate actions across enterprise applications and document repositories.

    Speaker: Vijai Gandikota, Principal Product Manager, AWS

    Time: 2:00 PM - 2:30 PM
    

    ---

    Simplify generative AI app development with Agents for Amazon Bedrock
    

    FMs effectively conduct conversations, create content, and drive productivity, but they can deliver more value if they’re equipped to transact with company systems to complete multi-step tasks. Fully managed Agents for Amazon Bedrock enables generative AI applications to help perform multi-step tasks across company systems and data sources, saving you from engineering prompts, training models, or manually connecting systems. Agents can plan and perform most tasks, from answering customer questions about product availability to checking inventory. Learn how to create agents by securely orchestrating enterprise APIs and data using natural language.

    Speaker: Santosh Ameti, Senior Software Development Manager, Amazon Bedrock, AWS  

    Time: 2:30 PM - 3:00 PM
    

    ---

    Architecture patterns for building generative AI applications
    

    In this session, we will talk about different architecture pattern for common generative AI use cases like text generation, summarisation, search and chat. We will also talk about how you can use embedding models along with vector database to search your document that is available in different data stores in your organization.
    

    Speaker: Praveen Jayakumar, Head of ML Solutions Architecture, AWS India
    

    Time: 3:00 PM - 3:30 PM
    

    ---

    Gain impactful insights with no-code AI & ML and generative BI
    

    Generative BI is an approach to Business Intelligence (BI) that uses Artificial Intelligence (AI) to make data more accessible and actionable. With generative BI, you can quickly and easily identify, surface, and leverage meaningful insights from data, even if you have no coding experience. In this session, cut through the hype and get to know what is real, what is available now, and how it will change the way organizations use data to make decisions. Learn how to use Generative BI will drastically increase productivity and allow companies to make better decisions, while greatly increasing availability to users.

    Speaker: Michael Armentano, APJ GTM Lead, Generative BI and QuickSight, AWS
    

    Time: 3:30 PM - 4:00 PM
    

    ---

    Deploy foundation models with performant, scalable, cost-effective inference on Amazon SageMaker

    Foundation models (FMs) enable cutting-edge AI applications, but deploying them at scale can be challenging. This session will explore strategies for performant, cost-optimized inference using Amazon SageMaker. We will provide an overview of new SageMaker features and tools to efficiently deploy hundreds of FMs for production inference. Attendees will learn how to architect scalable solutions that maximize performance and minimize costs for large-scale FM inferencing. Attendees will leave equipped with knowledge to unlock the potential of foundation models using SageMaker's performant, cost-effective, and scalable inference capabilities.

    Speaker:
    Dhawal Patel, WW ML Solutions Architecture Leader, AWS
    

    Time: 4:00 PM - 4:30 PM
    

    ---

    Accelerate foundation model development with Amazon SageMaker JumpStart
    

    Generative AI is empowering organizations to innovate faster to reinvent customer experiences and applications. But how can you ensure flexibility and control within your organization for generative AI application development? Amazon SageMaker provides managed infrastructure and tools to accelerate scalable, reliable, and secure foundation model (FM) development. In this session, dive into how Amazon SageMaker JumpStart makes it easy for ML practitioners to access FMs—including the best-performing publicly available FMs. Then see how you can evaluate, fine-tune, and deploy FMs using SageMaker, helping you to quickly get started.

    Speaker:
    Hasan Poonawala, Senior AI/ML Specialist Solutions Architect, AWS
    

    Time: 4:30 PM - 5:00 PM
    

  • Track 2: AI & ML stories

  • Leveraging GenAI for assisting life insurance sales agents on the field
    

    Sales Genie is a generative AI (artificial intelligence) based virtual life insurance trainer bot solution. This virtual life insurance trainer bot provides real-time, on-demand sales intelligence assistance to field sales agents, through a conversational chat platform. It improves sales productivity by providing the sales agents with appropriate answers to any sales related questions, financial market intelligence, and customer sales pitches. The solution leverages several generative AI models, along with associated transformer models (instruction tuned LLMs, vector embedding models, and speech transcription models). These are trained and contextualized, for specific Max Life Insurance sales processes, insurance products, market insights; and to handle all types of sales scenarios using Max Life Insurance sales training and product documents. In this session learn about the powerful capabilities of this solution.
    
    

    Speakers:
    Senthil Jeyachandran, AI & ML Specialist, Solutions Architect, AWS India
    Sanjay Thawakar, SVP & Head of AI Works & BPMA, Max Life Insurance
    

    Time: 1:30 PM  - 2:00 PM
    

    ---

    SirionLabs revolutionizes enterprise CLM using generative AI
    

    SirionLabs provides CLM (contract lifecycle management) solutions to enterprises. They work with multiple contracts and have multiple generative AI use cases, such as summarization of contract, contract renewal with updated clauses, tracking contractual SLAs (service level agreements) post signing, and more. SirionLabs tried several other LLMs, but were unable to achieve the right balance of performance and cost. Eventually they fine-tuned Llama 2 with their own data on Amazon SageMaker JumpStart, which helped them achieve the desired results in a cost-effective manner. Join this session to find out how AWS is helping SirionLabs optimize their inference/training ratio for a better price performance and scale.
    

    Speakers:
    Sujan Balachandran, Vice President of Products, SirionLabs
    Raviraj Prajapat, Principal Data Scientist, SirionLabs

    Time: 2:00 PM - 2:30 PM
    

    ---

    Staqu's JARVIS AI - Enhancing Homeland security and boosting retail with AI-powered analytics model
    
    JARVIS is an AI-powered audio-video analytics model developed by Staqu Technologies. It has the capability to transform CCTV footage into meaningful information and provides real-time alerts. JARVIS has helped law enforcement agencies, retailers, and manufacturers; with insights into real-time occupancy, efficient queue management, and footfall analytics. JARVIS’s customers include police departments of nine Indian states (Uttar Pradesh, Rajasthan, Bihar, Uttarakhand Punjab, and Haryana, and more), Adani Power, Croma, Starbucks, Café Coffee Day and WeWork. Join this session to understand how JARVIS can offer your organization innovative solutions for growth and security.
    

    Speakers:
    Varun Kumar, Solutions Architect, AWS India
    Anurag Saini, Co-founder & CTO, Staqu Technologies

    Time: 2:30 PM - 3:00 PM
    

    ---

    Using generative AI to make retail search engines smarter and more efficient
    

    Ace Turtle, a leading tech-native retail company is spearheading the retail industry’s next phase of transformation. It is vertically integrated to include design, local manufacturing, and marketing. Its proprietary technology uses data science from design to fulfilment, to meet ever-evolving consumer expectations. Ace Turtle is the exclusive licensee of global brands Lee®, Wrangler®, Toys”R”Us®, Babies”R”Us® and Dockers®, for India and other South Asian markets. The search within Toys”R”Us® and Babies”R”Us® websites was not yielding accurate results due to a lack of quality metadata and product descriptions, resulting in discovery challenges for potential buyers and affecting sales. In this session, learn how Ace Turtle built a search engine where product metadata is extracted through computer vision models from product images. The search capabilities were improved by adapting semantic search using VectorDB and Amazon Titan Embeddings.
    

    Speakers:
    Ashoka Gajula, Solutions Architect, AWS India
    Amit Kriplani, CTO, Ace Turtle

    Time: 3:00 PM - 3:30 PM
    

    ---

    Scaling enterprise AI with Kore.ai's generative AI platform
    

    Generic AI solutions often fail to address industry-specific challenges. LLMs provide a flexible foundation for tailored enterprise AI applications, but developing and deploying custom LLMs poses challenges around data privacy, infrastructure scaling, multi-tenancy, and cost optimization. In this session, explore how Kore.ai's generative AI platform enables enterprises to efficiently build, host, and scale customized LLMs on GPU clusters. Learn about real-world examples of LLMs fine-tuned on Falcon, OPT, and T5 to address unique client needs across industries. Discover how Kore.ai leverages Amazon SageMaker and Amazon Bedrock for fine-tuning and hosting custom enterprise LLMs. Get insights into how they observed performance gains by running LLMs on Amazon Elastic Compute Cloud Inf2 instances using AWS Neuron SDK.

    Speakers:
    Jaydev Nath, Solutions Architect, AWS India
    Prasanna Arikala, CTO, Kore.ai
    

    Time: 3:30 PM - 4:00 PM
    

    ---

    Blend app's inference at scale in a cost effective way using Amazon SageMaker multi-model endpoints
    

    Discover how Blend powers its AI image editing app for ecommerce sellers with its fine tuned foundational models, and is able to reduce the cost of serving upto 30% by using Amazon SageMaker multi-model endpoints. In this session, discover design considerations to train models in a cost effective way as well as inferencing at scale on Amazon SageMaker.
    

    Speakers:
    Krishna Chakka, Senior Solutions Architect, AWS India
    Jamsheed Kamardeen, Co-founder and CTO, Blendnow.com
    

    Time: 4:00 PM - 4:30 PM
    

    ---

    Travelopia uses Amazon Bedrock to improve efficiency of its sales agents
    

    Travelopia, a provider of unique travel experiences, needed a platform for its sales agents such that they could work directly with customers on call. They had a huge repository of data which could help their sales agents plan trips efficiently for customers. They built a travel assistant chatbot for the purpose, using Claude on Amazon Bedrock, agents for Amazon Bedrock, and LangChain. The chatbot provides sales agents with the information and tools they need to create on-the-fly itineraries, find the best dates and places to visit, and book hotels. It helped Travelopia's sales agents serve more customers. Join this session to find out how the travel assistant chatbot was developed.
    

    Speakers:
    Niyati Upadhyay, Solutions Architect, AWS India
    Sreekandh Balakrishnan, Technology Director, Travelopia

    Time: 4:30 PM - 5:00 PM
    

  • Track 3: AI & ML for business applications

  • Harnessing AI & ML for a unified 360 degree customer experience
    

    Salesforce Genie unifies customer data from marketing, and sales and service, to provide a 360-degree customer experience in real time. It uses Amazon SageMaker to integrate various customer platforms by enabling customers to use their own models, while also providing prebuilt and trained AI & ML models. The adoption of Salesforce Genie has been a gamechanger for Salesforce and customized versions across industries (healthcare, manufacturing, retail, and more), leading to increased usage. Join this session to learn how Salesforce Genie has transformed customer experience.
    

    Speakers:
    Upendra V, Senior Solutions Architect, AWS India
    Adithya Nittor Vishwanath, VP of Software Engineering, Salesforce

    Time: 1:30 PM - 2:00 PM
    

    ---

    Ecom Express develops LLM to improve location intelligence using AWS AI & ML
    

    Accurate location and address intelligence is essential for e-commerce logistics, but it is a challenge in India where addresses are not standardized and geographical directories are unavailable. Ecom Express used AWS AI & ML services to develop large language models (LLMs) for the address and geospatial domain, trained on 1.6 billion data points collected over the past 10 years (including over 1 million addresses and their geocoordinates from daily operations). In this session, learn how Ecom Express used Amazon Elastic MapReduce to collate and transform the address data, and Amazon SageMaker to train the AddressGPT model using GPU (graphics processing unit) accelerated instances. The new system reduced misroutes from 7% to 3% on the daily load of over 1 million shipments, resulting in cost and TAT (turnaround time) breach savings.
    

    Speakers:
    Dipta Shekhar Bhattacharya, Senior Manager, Solutions Architect, AWS India
    Manjeet Dahiya, VP & Head, Machine Learning & Data Sciences, Ecom Express

    Time: 2:00 PM - 2:30 PM

    ---

    Genie - AI powered enterprise architect assistant
    

    Genie, from Sixth Force Solutions is an AI-powered architecture assistant that improves the efficiency of your architecture practices, by assisting you in tasks such as model insights, validation, and documentation; among others. Genie’s use cases include Summary Assistant that produces a clear and concise summary of any conversation thread, Documentation Assistant that leverages generative AI technology to generate documentation, and AI assisted Model Search that helps you interact with your models and search for insights using natural language queries. Join this session to learn how AI powered architecture assistants can transform your business.
    

    Speakers:
    Praveen Hosur Narayana Gupta, Solutions Architect, AWS India
    Nizam Mohamed, CEO, Sixth Force Solutions Pvt Ltd

    Time: 2:30 PM - 3:00 PM
    

    ---

    Axio reinvents infinitely-scalable BNPL (buy now, pay later) for a faster shopping experience
    

    Axio, an NBFC (non-banking financial company) that services over 90% of Amazon.in's BNPL (buy now, pay later) transactions, needed a scalable machine learning (ML) engine to make sub-second BNPL credit decisions for shoppers, who present minimal information and no documents. Delays affect cart-checkout rates leading to a direct business loss. Axio leveraged Amazon SageMaker across the MLOps lifecycle, from data preparation to model training, testing, deployment, and retraining. They used Amazon SageMaker Notebooks to improve the productivity of their ML teams and Amazon SageMaker Data Wrangler to build feature pipelines and infer credit decisions in real time at scale, even during peak sales. Join this session to learn how AWS solutions helped Axio make accurate underwriting decisions, reduce their credit-decisioning latency to less than a second, improve cart-checkout rates, and support shoppers with BNPL purchasing options.
    

    Speakers:
    Deepesh Dhapola, Senior Solutions Architect, AWS India
    Sanket Tilekar, Vice President, Data Science, axio
    

    Time: 3:00 PM - 3:30 PM
    

    ---

    BigBasket creates AI-enabled checkout using computer vision algorithms from AWS

    BigBasket was facing challenges with its in-house ML (machine learning) algorithms for computer vision use cases, such as object recognition for AI enabled checkout at its Fresho stores. It was costly and time-consuming to train the models frequently to adapt to new products; and the training cycle time was increasing with the number of SKUs (stock keeping units), delaying time to market. BigBasket chose Amazon SageMaker, after being unsuccessful with other training solutions. With SageMaker distributed training and Amazon FSx for Lustre, they were able to reduce model training time by 45% and roll out to production faster. Additionally, the business team can now train models themselves with a single click, as all the data is on AWS. In this session, find out how multiple solutions from AWS helped BigBasket.
    

    Speakers:
    Sudhanshu Hate, AI & ML Principal Specialist, Solutions Architect, AWS India
    Vijayananda Reddy, Sr. Engineering Manager, BigBasket

    Time: 3:30 PM - 4:00 PM

    ---

    How Centre for Cellular and Molecular Biology (CCMB) uses AWS to drive AI & ML research in genomics
    

    The Centre for Cellular and Molecular Biology (CCMB) is a premier scientific research organization dedicated to conducting research of genetic material, how it varies among populations and how this leads to disparities in human health and diseases. CCMB scientists use nanopore sequencing to study DNA methylation, as abnormal DNA methylation patterns have been associated with various diseases. Nanopore sequencing provides the ability to detect DNA modifications. In this session, learn how CCMB built a neural network model on AWS that can distinguish between electrical signals of a canonical base versus a modified base, train modified base models, and run inference. By utilizing GPU instances, they accelerated the model generation process from several days on on-premises CPU machines to 3-4 hours per dataset. This research has the potential to improve early detection of cancer and save lives.
    

    Speakers:
    Mainak Chakraborty, Solutions Architect, AWS India
    Onkar Vasanttrao Kulkarni, Research Scientist, Sowpati Lab, Centre for Cellular and Molecular Biology
    

    Time: 4:00 PM - 4:30 PM
    

    ---

    Phenom People uses generative AI to build a smart HR ecosystem
    

    Phenom People, a global HR technology company was facing challenges scaling open-source LLMs for generative AI use cases, while needing to address cost control, security, language moderation, and latency concerns. To overcome these challenges, Phenom People created a custom LLM forked from the Llama-2-7B model and fine-tuned it on 50 billion tokens using Amazon SageMaker p4d instances. They used MosaicML StreamingDataset to stream data in shards from Amazon Simple Storage Service. This custom LLM enables a variety of use cases, including context-based email generation, context-aware FAQs for recruitment queries and response generation, and dynamic profile recommendation. In this session, learn how Phenom People achieved over 95% accuracy in search, over 90% accuracy in personalization, and improved job matching using AI.

    Speakers:
    Soujanya Konka, Senior Solutions Architect, AWS India
    Sekhar Jami, Engineering Manager, Phenom

    Time: 4:30 PM - 5:00 PM
    

  • Track 4: AI and ML solutions

  • Redefining possibilities with generative AI
    

    Step into a realm of innovation with LUMIQ, where generative AI isn't just a concept, but a revolution in action. The session will talk about how we’re pushing boundaries of existing AI/ML use cases and creating groundbreaking solutions across diverse domains - Marketing, Sales, Finance, HR, Customer Service, R&D, Legal & Compliance, etc. Additionally, we also delve into design thinking which covers development to deployment, all engineered for maximum and immediate ROI from generative AI initiatives. Step into a world where advanced technology meets practical results.
    
    

    Speaker:
    Vaibhav Dobriyal, Co-founder & Chief Product Officer, Lumiq
    

    Time: 1:30 PM  - 2:00 PM
    

    ---

    AI, without the angst 
    

    According to a recent Salesforce research, about 61% of employees use or plan to use generative AI at work, nearly 60% of those who plan to use this technology don’t know how to do so using trusted data sources or while ensuring sensitive data is kept secure. Attend this session to learn about Salesforce’s approach and principles that address the most pressing issue facing enterprises today in adopting generative AI at work.
    

    Speaker:
    Sridhar Hariharasubramanian, Senior Director, Salesforce India
    

    Time: 2:00 PM - 2:30 PM
    

    ---

    Reimagine your generative AI application with Redis Enterprise and Amazon Bedrock
    
    Redis Enterprise is not only a caching solution but can also work as a full-blown vector database, thereby providing real-time responses for requests and other features. Amazon Bedrock has a product-level integration with Redis Enterprise Cloud and in this session, we will see how we can build a generative AI app with Amazon Bedrock and Redis Enterprise as a real-time platform.
    

    Speaker:
    Abhishek Srivastava, Senior Partner Solution Architect, Redis
    

    Time: 2:30 PM - 3:00 PM
    

    ---

    Real-time analytics and multimodal data integration in RAG powered by MongoDB Atlas
    

    Embark on a journey into AI innovation with our session, delving into the seamless integration of MongoDB Atlas with RAG for machine learning. Uncover the dynamic management of multimodal data storage and real-time accessibility within the RAG framework. Discover the strategic advantages of harnessing the MongoDB Atlas Developer platform, empowering you to construct resilient AI/ML pipelines. Join us as we demystify the intricacies of this powerful collaboration, propelling your capabilities in the realm of artificial intelligence.

    Speaker:
    Ashwin Gangadhar, Senior Solution Architect, MongoDB
    

    Time: 3:00 PM - 3:30 PM
    

    ---

    Generating value from generative AI
    

    Business transformation and value creation through generative AI. As organizations try to crack the code on generating value with generative AI, hear more insights into how AI investments have driven value to date and what their future potential could be.
    Sanghamitra Pati, Leader, Strategy and Analytics (AI and Data), Consulting, Deloitte US-India Offices.
    

    Speaker:
    Sanghamitra Pati, Leader, Strategy and Analytics (AI and Data), Consulting, Deloitte US-India Offices
    

    Time: 3:30 PM - 4:00 PM
    

    ---

    Unlock your enterprise data's potential with intelligent search and generative AI
    

    The rapid growth of data has outpaced traditional search solutions. Information is scattered across systems, making it hard to find and use. Generative AI (GenAI) and large language models (LLMs) are transforming the way enterprises are able to solve traditionally complex challenges related to natural language processing and understanding. Join this session to learn how you can create state-of-the-art generative AI applications providing conversational experiences, by using Amazon Kendra with LLMs to ensure responses to users’ requests are based on your enterprise content.
    

    Speaker:
    Prem Ranga, Principal AI Specialist Solutions Architect, AWS
    

    Time: 4:00 PM - 4:30 PM
    

    ---

    Accelerating generative AI/ML adoption with AWS Storage
    

    Learn how you can accelerate your generative AI/ML journey by building a modern data architecture with Amazon S3. We’ll cover tips and best practices for implementing this architecture including incorporating 2023 re:Invent releases and show how you can quickly and cost effectively utilize your data to drive custom generative AI/ML outcomes.
    

    Speaker:
    John Mallory, Principal WW Storage GTM Specialist, AWS

    Time: 4:30 PM - 5:00 PM