QBuilder

Once upon a time in a bustling tech company, two junior full stack engineers, Lola and Susie, found themselves inspired by the challenges faced by their sales team in the application migration process from on-prem to Azure Cloud. They noticed that sales representatives often struggled to ask all the right questions during customer conversations, leading to delays and inaccuracies in cost estimation and project timelines.

Driven by their passion for problem-solving, Lola and Susie decided to embark on a journey to create a SaaS solution that would streamline the sales cycle and empower sales teams with the missing questions and accurate estimates they needed.

The Spark of Inspiration

Lola and Susie attended a company-wide meeting where the sales team shared their pain points about the application migration process. They were inspired by the opportunity to leverage their technical skills to create a tool that could bridge the gap between sales and engineering. They envisioned a solution that could analyze sales conversations and generate valuable insights to improve the sales cycle and timeline/cost estimation process.

The Quest for Knowledge

To bring their vision to life, Lola and Susie delved deep into researching Natural Language Processing (NLP) techniques. They studied various algorithms and models that could analyze audio transcriptions, emails, and notes to identify missing questions. They also explored sales cycle and project management methodologies to ensure accurate timeline and cost estimation.

During their research, Lola and Susie learned about Natural Language Processing and ChatGPT . They discovered how these technologies could be applied to sales conversations, enabling the system to understand the context and extract essential information.

The Building Blocks

Equipped with newfound knowledge, Lola and Susie set out to build QBuild: A Sales Conversation Analyzer. They divided their work into two main components: question identification and timeline/cost estimation generation.

For question identification, they developed a robust NLP pipeline that could process and analyze various data sources. By parsing conversations, they developed the system to identify gaps in the sales conversation, highlighting the missing questions that sales reps needed to ask customers. They combined front-end and back-end technologies to summarize conversations, break them into their respective pieces and parse them into ChatGPT prompts.

To generate accurate timelines and cost estimates, Lola and Susie integrated their system with project management tools. They collaborated with engineering managers and directors to define the necessary parameters and factors that influenced the cost and estimation process. By storing historical data and parsing conversations they designed an algorithm that could produce reliable estimates based on the provided information.

Overcoming Challenges

Throughout their journey, Lola and Susie faced numerous challenges. They encountered issues with data inconsistency, which required extensive preprocessing and data cleaning. They also had to tackle the complexity of understanding customer needs and matching them with technical requirements. They had to carefully design APIs and ensure seamless data flow between the different technological components.

Triumph and Transformation

After 24 hours of hard work, Lola and Susie successfully built QBuild. They conducted thorough testing, refining the model and algorithms to achieve optimal performance. The solution was well-received during internal demos, and the sales team expressed their gratitude for the newfound efficiency it brought to their workflows.

Lola and Susie's journey not only resulted in a valuable tool but also transformed their own skills and understanding of the application migration and building process. They grew more adept at working with NLP techniques, project management methodologies, and cross-functional collaboration.

Their project served as a testament to the power of innovation and collaboration, reminding them that even as junior engineers, they could make a meaningful impact within their organization.