Provectus at AWS re:Invent 2023: Leading the Charge in Generative AI Innovation

Author:
Provectus, AI-first consultancy and solutions provider

AWS re:Invent 2023 event was more than just a conference — it was a celebration of the latest in cloud technology and AI innovation. The event, known for its deep dives into new tech and hands-on experiences, brought together the brightest minds in the field.

Attracting over 50,000 attendees, the event offered hundreds of keynotes, leadership sessions, breakout tracks, and Learn & Play activities. For those unable to attend in person, the option to join virtually provided a great opportunity to enhance skills and gain a deeper understanding of AWS technologies.

As a recognized AWS Premier Consulting Partner, Provectus was excited to sponsor re:Invent and share our latest breakthrough: a Generative AI-powered solution for insurance underwriting. This innovation, at the cutting edge of Gen AI, demonstrates our commitment to pushing boundaries and delivering real-world applications that offer tangible benefits to business.

But there was more going on at AWS re:Invent 2023 than our Generative AI showcase. We would like to share an overview of our experiences: from the keynotes we attended to our tech demos and interactions with clients and partners.

Overview of re:Invent ‘23

AWS re:Invent is one of the world’s most significant technology events. It brings together thinkers, builders, and learners, creating a fertile environment that sparks creative ideas, initiates new projects, and scales innovation.

This year’s re:Invent was no exception, but with a twist: the spotlight was not only on cloud technology, but also on AI & Data. Given the remarkable advancements in Generative AI over the past year, this shift in focus was hardly surprising. A growing interest in Gen AI was evident across the event’s keynotes, leadership sessions, breakout and deep-dive sessions, and bootcamps.

But re:Invent is not just about advancing technology; it is also about having fun. This year’s event was a well-balanced mix of tech and entertainment. The Expo wowed attendees with spectacular displays of race cars and robots, adding an exciting interactive dimension. A notable highlight was Major Lazer’s performance of a David Bowie cover at the re:Play party.

As always, keynotes were the cornerstone of re:Invent, featuring five stellar presenters:

1. Keynote with Peter DeSantis
2. Keynote with Adam Selipsky
3. Keynote with Swami Sivasubramanian
4. Keynote with Ruba Borno
5. Keynote with Dr. Werner Vogels

We highly recommend watching all five keynotes, but if your time is limited, make sure to catch keynotes #2, #3, and #5. Here’s why:

• In his keynote, Adam Selipsky outlined an ambitious vision for AWS. It includes the expansion and diversification of AI models, services, and use cases; AWS’s strategic partnership with NVIDIA to develop and use more powerful GPUs; and leveraging data as a crucial business differentiator.
• Swami Sivasubramanian continued the AI + Data narrative with his assertion that humans will coexist with Generative AI and Data to collaboratively drive creativity, fostering a reinforcement cycle. A key point was the concept that ‘no one model will rule them all’, aligning with AWS’s strategy of expansion and diversification.
• In his “Matrix” keynote, Werner Vogels touched on two themes: architecting with cost in mind, and the value of traditional AI. He advised engineers to consider cost as a factor in sustainable development and cautioned against getting swept up in the Gen AI hype. Instead, he recommended focusing on more traditional AI use cases and applications.

In alignment with the AI + Data focus, AWS introduced several new services and announced significant enhancements to existing ones. Most notably:

• Next-Generation AWS Chips: AWS Graviton4 and AWS Trainium2 will continue to enhance performance and efficiency in running Generative AI and other workloads.
• Amazon Q: A Generative AI-powered enterprise chat assistant, enhancing productivity with tasks like answering questions, discovering knowledge, and drafting documents. The service is customized with organizations’ own corporate data, taking intelligent work to the next level.
• Amazon Q in Connect: Integration to assist customer service agents with responses and actions. (Also available in QuickSight, CodeWhisperer, Connect, Supply Chain).
• Expanded Model Choices in Amazon Bedrock: A wider range of models for Generative AI applications, including new additions from various AI companies (Anthropic’s Claude 2.1 and Meta Llama 2 70B). New capabilities for comparing and selecting the right model for an organization’s specific use case.
• Model Customization in Amazon Bedrock: A capability enabling organizations to customize foundation models, which are available in Amazon Bedrock, with their own corporate data (RAG methods).
• Improved Control and Orchestration for Bedrock Agents: Enhanced capabilities that include improved control of the orchestration and better visibility into the chain of thought reasoning, to further accelerate the development of Gen AI applications.
• Guardrails for Amazon Bedrock: A capability that enables organizations to deliver relevant and safe user experiences aligned with their company policies and principles, and protect users against harmful content.
• New Amazon SageMaker Capabilities: Five new capabilities to simplify and make more cost-efficient the process of building, training, and deploying models for generative AI.
• Amazon SageMaker HyperPod: A new purpose-built infrastructure for distributed training of large language models. It solves the hardware problem for developers, allowing them to train large models faster and more efficiently, with up to 40% improvement in training performance.
• AWS Clean Rooms ML: A privacy-preserving service enabling the deployment of AI/Gen AI models for company collaborations without sharing proprietary data.
• Amazon Neptune Analytics: A new analytics database engine that combines graph and vector databases, to analyze graph data or data lakes on Amazon S3 storage, at scale. Neptune Analytics speeds up analytics including vector search by 80x.
• AWS Titan Image Generator: A new service using multimodal embeddings can create images from text descriptions, or customizing existing ones. The service offers a wide range of image creation and modification capabilities.
• New Capabilities for AWS Supply Chain: Introduction of four new features to improve forecasting, product replenishment, and supplier communication; new Generative AI-powered features.
• AWS Serverless Innovations: New offerings for Amazon Aurora, Amazon Redshift, and Amazon ElastiCache, enhancing data analytics and data management capabilities.

We encourage you to watch the keynotes and explore the announcements for yourself. You can watch all the conference content on demand so you won’t miss a minute of this incredible learning opportunity. All available re:Invent content can be found here.

Innovating with Intent: Provectus Brings Gen AI to re:Invent

Provectus is an AI-first consultancy and solutions provider, helping organizations transform through AI. Provectus leverages the power of cloud, data, and AI to reimagine the way businesses operate, compete, and deliver customer value.

As we headed to re:Invent, we expected Generative AI to dominate discussions — and it did. The event was buzzing with vendors showcasing their chatbots, personal assistants, and “enterprise-ready” Gen AI platforms.

What sets Provectus apart and positions us for success is our unique combination of strengths:

1. Provectus stands out as the only pure-play, hands-on data and AI consulting partner. Unlike others who previously dabbled in areas like cloud migration, SAP, Windows, Salesforce, and Business Process Outsourcing (BPO), and only recently shifted focus to AI, our expertise has always been firmly rooted in AI + Data.
2. Provectus boasts the most comprehensive and in-depth suite of AI solutions in the industry. We address specific, concrete niche use cases for sectors like BFSI, HCLS, CPG, Retail, eCommerce, as well as general functions including Sales, Finance, Customer Success, and Supply Chain.
3. Our distinctive AI transformation methodology is designed for top-down implementation, starting from the Board level. Our approach includes C-level education and strategy, day-zero security, risk, governance, and compliance playbooks and infrastructure, as well as a proven methodology for use case discovery and prioritization.

At re:Invent, the Provectus team attended all keynotes and participated in various sessions led by industry experts. It quickly became evident that:

While Generative AI offers numerous benefits to businesses, it is Data that unlocks its full potential. Data acts as a key differentiator in both in-house and customer-focused Generative AI use cases. The synergy of AI, Data, and Human expertise can create a reinforcing cycle that drives business success.

At the same time, the market has clearly entered a maturity stage where generic expertise in AWS, Cloud, AI, Data, and software engineering alone is no longer sufficient for a business to stand out. We anticipate a growing demand for more specialized skills and solutions, and Provectus is poised to innovate and evolve in that direction.

To set the stage for more specialized and focused Gen AI + Data solutions, we showcased our own innovations at re:Invent.

We presented a groundbreaking demo of our Generative AI-powered Insurance Underwriting solution, developed on AWS (Amazon Bedrock) with an advanced Large Language Model (LLM) from Cohere, that can transform the traditional underwriting process. We demonstrated how our system automates the distillation of complex 100+ page technical documents into precise 2-page summaries in just minutes, a task that normally takes underwriters 7-10 days.

Our live demo highlighted the solution’s ability to quickly pinpoint critical information, including key points of interest, and assess positive and negative aspects and crucial KPIs. The audience saw firsthand how the consistency and superior quality of these automated reports led to substantial time savings and more efficient decision-making in underwriting.

Attendees at re:Invent had the opportunity to experience a live demo of the solution at Cohere’s booth #124 in the Infrastructure Zone, generating considerable excitement among business representatives from various industries.

Provectus also engaged with attendees at booth #1244 in re:Invent’s AI/ML & Data Zone. Over a hundred attendees visited our booth daily, where our technology leaders presented our Universe of Generative AI Solutions powered by Amazon Bedrock, and explained how Gen AI can boost operational efficiency, optimize costs, and accelerate business growth.

Beyond our Generative AI showcase, the Provectus team introduced our S&OP and CLO Surveyor solutions, and discussed AI adoption and transformation strategies. We also shared our vision of Generative AI.

As always, re:Invent was more than just a work event – Provectus hosted a series of entertainments for our clients and partners.

• At Topgolf, located in the MGM Grand Hotel and Casino, guests tested their golf skills in our exclusive bays, competed in our Longest Drive contest, and indulged in Topgolf’s signature food and drinks, all while discussing the transformative potential of Generative AI.
• Our Generative AI Executive Dinner at Stanton Social Prime in Caesars Palace offered an elegant setting for showcasing our partnership with Cohere, and a chance for industry leaders to mingle and delve into real-world AI applications.
• Finally, at the landmark Smith & Wollensky in the Venetian, our Executive Dinner provided a unique opportunity for guests to engage with Provectus leaders over signature cocktails and fine dining while exploring AI’s real-world business applications, and learning how our partner programs can initiate their AI transformation journey.

AWS pulled out all the stops to entertain attendees too.

The newly introduced re:Invent Rec Center offered conference-goers a chance to participate in games inspired by top sports franchises, including a Formula One racing simulator and an NFL passing challenge. The Bundesliga free-kick challenge was a big hit, offering real-time analysis of kicking mechanics.

The re:Invent Expo was an interactive hub, abuzz with AWS Partners demonstrating their latest offerings. Attendees gathered for the excitement of the AWS DeepRacer League Championships, where machine learning met model car racing. The AWS Builders Fair showcased innovative projects, providing a unique mix of technology and creativity.

Capping off the event was the re:Play party, an unforgettable night of music, networking, and entertainment. This iconic party is renowned for its vibrant atmosphere, providing a perfect setting for attendees to unwind, celebrate their newly acquired knowledge, and build lasting connections.

Overall, AWS re:Invent 2023 provided attendees with an abundance of opportunities to socialize, engage in playful activities, and unwind in a relaxed atmosphere.

AI + Data = the Future of Humanity

Amazon Web Services places a strong emphasis on the synergy of AI and Data as pivotal to the future, envisioning a collaborative ecosystem where AI, Data, and Humans not only coexist, but mutually enrich one another.

This fusion of AI and Data is also seen as the bedrock of digital transformation for organizations. The event’s comprehensive lineup of keynotes, sessions, and tracks clearly demonstrate AWS’s ambitions for the ongoing technology boom.

This year’s re:Invent marked a significant milestone in the evolution of Generative AI. Its rapid advancement, coupled with the increased availability of new foundation models, more robust chips for training and inference, and a variety of on-demand Gen AI use cases, favorably positions businesses in various industries to achieve sustained, meaningful outcomes. This progress steers organizations away from short-term, quick-fix AI/ML solutions, setting a course for long-term success through AI transformation.

Provectus is thrilled to be a leader in the journey towards an AI-driven future. We firmly believe in the transformative power of AI and Data to create a better and safer world for everyone.

We look forward to continuing this work and sharing our progress in the Gen AI field at AWS re:Invent 2024!