How Boggl AI Transforms Your Product Lifecycle Management with Amazon Bedrock by Ensuring Data Security

By Swarna Hebbar, Founder and CEO – Boggl AI
By Bakrudeen K, Head AI/ML Practice – Shellkode
By Ganesh Sawhney, Sr Partner Solutions Architect – AWS
By Rahul Kumar, Cloud Sales Rep – AWS

In this blog post, we will explore how Boggl AI empowers product teams, including product managers, developers, and QA engineers to transform their end-to-end product lifecycle management using generative AI capabilities, powered by Amazon Bedrock.

Managing product lifecycles is often complex and challenging. From gathering customer feedback, to planning roadmaps, and preparing product documentation, these tasks requires time and precision. Boggle AI offers a voice-powered product management assistant for fast moving product teams. By leveraging the powerful capabilities of Amazon Bedrock, Boggl AI helps teams streamline this process with an automated, scalable solution for product management while ensuring that data remains secure and private.

Figure 1 illustrates the AI-powered product development lifecycle with Boggl AI. It begins with inputs such as customer feedback, meeting transcripts, voice or text Input, and brainstorm notes. Boggl AI triages this feedback and converts all the inputs into a structured roadmap and requirements, which then generate release notes and test cases.

Figure 1: AI-powered product development lifecycle with Boggl AI

Boggl AI is a voice-powered AI assistant that enables product teams to efficiently manage the entire product lifecycle, such as customer feature triage, roadmap planning, requirements, task management with teams, and customer documentation.

Consider a founder of a growing SaaS startup managing numerous feature requests monthly. Your product team is busy building new features, leaving little time to manage requests, update roadmaps, or create documentation. Boggl AI streamlines this process with voice commands, allowing generative AI to craft AI-driven roadmaps, automatically triage customer requests, and generate detailed requirements and product documentation, seamlessly in a click.

Automating Customer Feedback Triage

A critical aspect of any business is customer success. It is crucial to constantly hear customer feedback and incorporate it into the product development for a tech company. SaaS companies who can receive a few dozen to hundreds of requests/feedback each month often to sort them and assign them to product managers manually. Boggl AI uses advanced AI models to automatically triage and analyze customer feedback and requests to identify the most important insights. This empowers product teams to accurately identify what needs to be prioritized based on the expected impact.

Intelligent Roadmap – Focusing on What Matters Most

Building a product roadmap often requires balancing customer needs, technical feasibility, and business priorities. With the voice-powered AI assistant, teams can simply narrate their ideas, and Boggl AI will analyze the feedback and internal data to generate data-driven roadmaps. By automatically calculating a prioritization score based on each requirement’s reach, impact, confidence, and effort, product managers are equipped with a dynamic, easy-to-maintain roadmap. This approach allows them to efficiently manage growing requirements and customer requests with a scalable and effective method for prioritization.

Requirements and Task Breakdowns

Boggl AI’s voice-powered AI assistant simplifies the creation of key product lifecycle documents for users. By simply narrating requirements, teams can automatically generate product requirements, test cases, and task breakdowns. This eliminates the need for manual documentation, ensuring consistency and accuracy across teams. The requirements are linked with task breakdowns and help in streamlining workflows to allow product managers to focus on strategy while ensuring smooth execution.

Release Notes and Customer Product Documentation

Successful product releases require, well-organized communication. Boggl AI makes it easy for teams to automatically generate release notes and other customer-facing product documents as soon as a new feature is launched or a product update is deployed.

A key advantage is how well Boggl AI adapts the content to its audience, whether internal teams or end users. With a click, the platform creates tailored documentation that ensures everyone gets the right level of detail.

Mid-Sized SaaS Company’s Success with Boggl AI

A mid-sized SaaS company scaling operations adopted Boggl AI to automate product documentation, reducing manual effort and speeding up the creation of customer-facing materials like product guides and release notes. The results? They were able to release product documentation 60% faster than the time they were originally spending before adopting Boggl AI. This has also enabled them to delight their customers by releasing their product documents once a month which used to be once a quarter before. The company now plans to expand Boggl AI’s use for customer feedback triage and road mapping to further optimize feature prioritization and decision making.

Initial Data Security Challenges

Boggl AI initially relied on APIs from other LLMs to power its AI-driven features. Due to this, there were challenges in reassuring customers that their sensitive data would remain secure and would not be used for further training of the AI models. The use of direct APIs from third-party models posed a significant data security concern, as it was difficult to provide a clear guarantee that the data wouldn’t be reused for model training purposes. This lack of transparency and control over data handling created trust issues, especially for customers managing sensitive or proprietary information. Consequently, it became essential for Boggl AI to seek a more secure, privacy-focused solution.

The Value of Combining Boggl AI and Amazon Bedrock

To address the above challenges, Boggl AI infrastructure transitioned to Amazon Bedrock, ensuring that the customer data was used solely for inference and never for model training. This shift has empowered the Boggl AI team to build trust with their clients, especially those dealing with sensitive feedback or proprietary data.

Boggl AI’s use cases have consistently delivered high-quality content using the Claude 3.5 Sonnet model through Amazon Bedrock. Additionally, the nuanced contextual understanding and consistency provided a compelling reason to switch to the Claude 3.5 Sonnet model on Amazon Bedrock.

Overall, Amazon Bedrock proved essential for ensuring data security and delivering high-quality outputs, enabling us to provide a secure and enhanced product management experience for Boggl AI’s customers.

“99.9% of customers express confidence in Boggl AI’s data security protocols on AWS”

– Swarna Hebbar, Co-Founder – Boggl AI

Boggl AI and AWS Architecture – Built for Efficiency and Security

Figure 2 illustrates a workflow depicting Boggl AI using Amazon Elastic Compute Cloud (Amazon EC2) to power the application server and the custom logic that selects the most suitable LLM (large language model) for tasks like customer feedback triage, roadmap generation, or creating detailed requirements. Once the LLM is chosen, Python Celery triggers AWS Lambda functions to access Amazon Bedrock, enabling efficient and secure use of the most relevant AI models for each use case, ensuring high-quality outputs. Each step in the workflow is connected by arrows, showing the sequence of interactions between services.

Figure 2: Boggl AI and Amazon Bedrock Architecture

Architecture Overview

The following is an architecture overview illustrated in Figure 2, for the integration of Boggl triggering a FAST API hosted on EC2 to manage multiple model options (Claude 3, LLaMA 2, etc.) for prompt-based responses.

1. User request via Boggl interface, providing inputs like the prompt and selecting the preferred AI model based on the use-case.
2. Boggl sends a request to the FAST API running on an EC2 instance. The request payload also includes the desired model.
3. FAST API sends user’s prompt to the Amazon Bedrock Foundation model API.
4. Model processes the prompt and FAST API responds back to Boggl which in-turn presents the output to the user

Shellkode, an AWS partner, was instrumental in developing and implementing the solution. This task required an in-depth understanding of both the legacy and new systems to ensure a smooth transition without compromising the content generation process.

“Since using Amazon Bedrock, Boggl AI has had zero data leaks or unauthorized access incidents.”

– Swarna Hebbar, Co-Founder – Boggl AI

Impressive Data Security and Content Quality Metrics with Amazon Bedrock

• Claude 3.5 Sonnet generates content that’s 25% more accurate for specific use cases compared to other models.
• Enhanced content relevance has increased customer engagement by 35%.
• AWS and the third-party model providers on AWS do not use any inputs to or outputs from Amazon Bedrock to train Amazon Titan or any third-party models, ensuring a higher degree of data security and privacy with Amazon Bedrock.
• With secure data and high-quality content, over 90% of the prospects trust Boggl AI’s data security measures on Amazon Bedrock which has impacted in faster sales cycle and product adoption.

Conclusion

Boggl AI, powered by Amazon Bedrock, transforms product lifecycle management by streamlining processes from customer feedback to documentation while ensuring data security. Product teams can now automate time-consuming tasks, protect sensitive information, and focus on strategic development. This solution, developed in collaboration with AWS partner Shellkode, demonstrates how AI can enhance efficiency and security in product management. To explore how Boggl AI and Amazon Bedrock can benefit your organization, visit the Boggl AI website or contact Shellkode for implementation details

.

Shellkode – AWS Partner Spotlight

Shellkode is an AWS Partner that helps organizations harness the power of change, speed, and innovation to create all-around business value by putting data and cloud at the core of your solutions.

Contact Shellkode | Partner Overview