Leveraging GenAI to Empower JBS Dev’s Sales and Delivery Teams

How It Started

Two years ago, JBS Dev's internal landscape looked significantly different. With a small sales team in place, the lift of answering project-related queries fell upon our delivery team members. They were the keepers of knowledge, drawing on years of experience and a deep understanding of our company's history and past projects. However, as our company expanded and new sales personnel joined our team, a knowledge gap emerged. We had to find a way to share information quickly and efficiently with our new team members, seeking information about our past projects in specific industries or with particular clients.

To bridge this gap, we initially set up an internal chat channel, enabling the sales team to tap into the collective brainpower of the JBS Dev delivery team. While effective, this solution needed to be more scalable, and it covered only a fraction of our comprehensive 25-year project history. The need for a more efficient, less time-consuming solution was clear, leading to the start of our internal chatbot project.

Overcoming Challenges with Innovation

The Technical Labyrinth

JBS Dev Bot is built using a Retrieval Augmented Generation (RAG) architecture. A RAG architecture allows us to dynamically augment our prompts with relevant data. Developing our chatbot has started with the intricate process of data ingestion. Structuring decades' worth of data into a usable format while filtering out irrelevant or sensitive information was a significant task. Moreover, tuning the chatbot to understand and focus on the essence of the queries it would face required meticulous effort. We aimed to create a digital repository that could match the knowledge and insight of a 25-year veteran JBS Dev employee without the need for human intervention for general queries.

Ethical and Practical Considerations

Ensuring the chatbot provided accurate information without overstepping its bounds was another critical aspect of our development process. We used prompt engineering to tailor how JBS Dev Bot responds to questions and how to address the situations when it's unsure of the correct answer. This was crucial in maintaining the integrity and usefulness of the information provided.

Utilizing AI for Enhanced Efficiency and Insight

Implementing our chatbot has had a transformative impact on our internal operations. Not only has it streamlined the process of knowledge sharing among our sales and delivery teams, but it has also unearthed insights that were previously buried under the sheer volume of data we possess. The chatbot's ability to provide comprehensive answers, free from human biases, has been a significant benefit.

What’s Next?

Our journey doesn't end here. We are committed to continuously improving the chatbot, ensuring it remains up to date with the latest information and can serve even more functions within our organization. The potential applications are vast, from generating support documentation to project staffing requirements and improving resource management through machine learning analysis of personnel skills and project requirements. The cornerstone of any project's success lies in the effective allocation of resources. Recognizing this, we are exploring the use of machine learning to revolutionize how we manage our talent pool. Our AI systems can identify patterns and insights that human analysis could overlook by analyzing vast datasets, including personnel skills, past project involvements, and project outcomes.

Furthermore, we are exploring ways to use generative AI to enhance employee access to essential company policies and resources, making the user experience as seamless and informative as possible. The goal is to make every interaction with our internal systems as intuitive and informative as possible, reducing employees' time searching for information and allowing them more time to focus on their core responsibilities.

Conclusion

Developing an internal AI-powered chatbot has been a journey of innovation, challenge, and discovery. It has improved our operational efficiency and opened new avenues for applying AI within our company. As we continue to iterate and expand its capabilities, we are excited about the potential of AI to transform the way we work and share knowledge within JBS Dev. We aim to improve our team’s efficiency and productivity and foster a more engaging and supportive work environment. By harnessing the power of AI and machine learning, we are setting the stage for a future where our operations are more agile, our teams more empowered, and our projects even more successful.