How often do you work on projects with limited or missing documentation? This lack of clear information can create a ripple effect of misunderstandings, delays, and rework. That’s why clear communication is essential.
To address this challenge, we believe that large language models (LLMs) can provide relevant answers to business questions directly in the code itself. Many project roles, including business analysts (BA), developers, quality assurance (QA) specialists, Project managers (PM), and product owners (PO) can benefit from this potential solution.
CodeTeller is an innovative solution that bridges the gap between code and clear business understanding by leveraging the power of large language models (LLMs). It treats your code as the single source of truth, delivering answers and insights directly within your chat platform.
CodeTeller employs a robust architecture that combines several advanced technologies like Retrieval-augmented Generation (RAG), Microsoft Azure AI, and Semantic Kernel. RAG capabilities enhance the LLM by retrieving facts from an external knowledge base, supplying the model with the most accurate and up-to-date information. This ensures that the generated responses are not only contextually rich but also reflect the latest insights.
CodeTeller utilizes the power of Azure AI Search, an information retrieval platform that excels at understanding the meaning behind your code. This seamless integration empowers CodeTeller to translate code into clear business language, making it easier for various stakeholders across your organization to comprehend and collaborate effectively.
Furthermore, CodeTeller uses the Semantic Kernel, an open-source software development kit (SDK) designed to combine AI services effortlessly. Within CodeTeller’s architecture, Semantic Kernel acts as an orchestrator, automatically coordinating plugins that enable the LLM to achieve project goals step by step. By harnessing the power of multiple AI models and session context, these plugins create sophisticated pipelines that enable CodeTeller to automate complex tasks for users throughout their project management sessions, significantly improving overall efficiency.
CodeTeller is already up and running, providing valuable insights to users. First Line Software actively working to improve its capabilities through developing new plugins and implementing evaluations. In the future, we plan to integrate CodeTeller with Confluence, enabling the consolidation of all current project information. This integration aims to streamline requirements investigation and testing, facilitating fast and high-quality deliveries.
CodeTeller represents a significant step forward in the world of project management. By harnessing the power of AI, CodeTeller empowers teams to work more efficiently, collaborate effectively, and unlock the hidden potential within their code.
Ready to experience the CodeTeller difference? Contact us today to learn more about how CodeTeller can revolutionize your project management approach.
Business and System Analyst at First Line Software.
Ksenia Dziuba brings more than three years of experience in system design and requirements documentation. Her role involves identifying what businesses need for better integration, ensuring data flows smoothly between systems through careful planning, and working alongside technical teams to ensure everything works well together. Ksenia’s work is key to making sure business goals and technology go hand in hand.
Whether you have a problem that needs solving or a great idea you’d like to explore, our team is always on hand to help you.
First Line Software is a premier provider of software engineering, software enablement, and digital transformation services. Headquartered in Cambridge, Massachusetts, the global staff of 450 technical experts serve clients across North America, Europe, Asia, and Australia.
Clinovera