The development of Pulse: From proof of concept to generative AI accelerator
Pulse started unexpectedly – built in a tent within hours of Amazon Bedrock’s release. In this article, we’ll explore how Pulse evolved from a quick proof of concept into a generative AI accelerator that is now helping businesses streamline workflows and achieve faster results.
Where it all began
Back in July 2022, when the Amazon Bedrock preview was first released, the team at Firemind were out camping in Snowdon. Managing Director, Ahmed Nuaman, received news that they now had access to Amazon Bedrock and within just 24 hours, he had already built a proof-of-concept using the new service.
Ahmed recalls, “I got a message on my phone that we had access to Bedrock. I stepped into my tent, and half an hour later I had already built something to showcase why Chelsea were having a bad season.” This early version featured a basic interface that analysed the football club’s performance and provided a summary of why they were underperforming.
This initial success kickstarted the development of Pulse – Firemind’s generative AI tool built on Amazon Bedrock. What started as a simple proof-of-concept quickly expanded beyond a basic demonstration into a versatile tool capable of handling more complex workloads and extracting insights from diverse data sources.
Bringing Bedrock to life with Pulse
As Amazon Web Services (AWS), began promoting the benefits of Amazon Bedrock, Firemind saw an opportunity to showcase its practical use. Pulse became a way for Firemind, AWS, and customers to quickly demonstrate the value of generative AI.
Pulse provides a user-friendly interface that enables users to upload their own datasets and create custom prompts. With access to a range of language models, customers can use generative AI to automate various tasks, such as analysing customer reviews or summarising key information from annual reports or other documents.
Pulse also supports more advanced workflows with its batch processing capabilities for Intelligent Document Processing (IDP), enabling the automation of complex, repetitive tasks at scale. These capabilities allowed companies to integrate generative AI directly into their operations, enhancing both productivity and accuracy.
“We were getting introduced to enterprise customers from the AWS team, and we would very quickly demonstrate the value of Pulse,” explains Ahmed. “We’d download publicly available data, like Trustpilot reviews, and very quickly, we could show the customer how they could automate those low-value tasks. Some companies were even employing people full-time to do this kind of work, so it was a real eye-opener for them.”
Showcasing Pulse to AWS sellers and industry experts
As Pulse evolved, Firemind began leveraging Pulse as a pre-sales tool to showcase the value of generative AI. The team attended events and held workshops, designed to teach AWS sellers and other experts how to use the Pulse tool to demonstrate the capabilities to their customers.
These workshops often followed a “Dragon’s Den” style format, allowing participants to experience the Pulse tool firsthand and explore potential use cases for their own businesses. This interactive approach allowed teams to see, in real-time, how Pulse could solve common business challenges, making the sessions both informative and engaging.
Pulse was developed in response to specific customer needs. Our team worked closely with customers and AWS to ensure that Pulse met real business demands. By working backwards from industry requirements and customer feedback, we made Pulse a customer-led product, designed to solve genuine challenges across sectors.
Over the course of two years, Pulse’s development has been shaped by ongoing customer input and AWS feedback. We continuously listened to suggestions and innovated to align with evolving customer needs. For instance, at the TINtech London Markets Insurance event, we showcased Pulse’s FNOL (First Notice of Loss) automation solution, specifically aimed at transforming insurance workflows. Similarly, at the AWS Media and Entertainment Symposium, we demonstrated how Pulse could be used for video analysis – illustrating its flexibility for different industries.
To demonstrate Pulse’s full potential, Firemind has developed a solutions catalogue that highlights a diverse range of real-world use cases.
Pulse today: Firemind’s generative AI accelerator
Pulse has now become Firemind’s popular generative AI acceleration tool, that rapidly showcases generative AI use cases to customers, clearing a path from initial adoption to productionised workloads. Pulse showcases the diverse applications of generative AI, significantly accelerating a customer’s time to value. It has become a crucial part of Firemind’s offering, helping the company and its customers unlock the full potential of generative AI.
A key feature of Pulse is its single-tenancy architecture, where each instance is deployed directly into the customer’s AWS environment. This approach was chosen to prioritise data security, ensuring that customer data remains fully isolated and under their control.
To further simplify deployment, the Firemind product team has integrated Pulse into the AWS Marketplace over the past three months. This integration allows both new and existing customers, as well as AWS partners, to easily deploy Pulse using the AWS self-service process. By being available on AWS Marketplace, Pulse is more accessible, offering reduced administrative overhead, increased visibility to AWS’s expansive customer base, and flexible pricing options.
Arc: complementing Pulse’s capabilities
While Pulse is a versatile generative AI tool, it doesn’t address every challenge faced during AI adoption. To complement Pulse, we developed Arc – a solution designed to handle the management complexities that arise as customers move AI projects from proof of concept to full production.
Arc augments our AIOps managed service by providing an intuitive dashboard that removes the heavy lifting of managing AI workloads. It streamlines AI workload management by focusing on proactive alerting, model evaluation, and optimisation, addressing major barriers to production. Arc ensures our customers are prepared for changes, such as model updates, new releases, or deprecations – making AI management straightforward and reducing risks in production.
To conclude
Pulse has grown from a quick proof of concept to a key solution for accelerating generative AI adoption. Developed in response to specific customer needs and ongoing AWS feedback, Pulse focuses on security, efficiency, and practical applications – empowering companies to integrate AI seamlessly and drive real impact across industries. Complementing Pulse, Arc helps customers manage and optimise their AI environments effectively as they transition into production.
To further support our customers and help them keep up with all the changes in the AI landscape, we created labs.firemind.com – a dedicated resource for publishing news and insights on generative AI, AIOps, modern data strategies, and more. Firemind is committed to helping our customers innovate and stay ahead.
Get in touch
Want to learn more?
Seen a specific case study or insight and want to learn more? Or thinking about your next project? Drop us a message!