Inspiration
As someone who has been coding for years, I’ve often found myself revisiting the same stages of the Software Development Life Cycle (SDLC), from planning and design to testing and deployment. Despite the availability of documentation and tools, navigating this process can be overwhelming and repetitive. I wanted to create a tool that not only guided developers through the SDLC but also made it easier to get help at each stage. Leveraging my experience and curiosity about agent-based systems, I saw the perfect opportunity to use Google ADK to bring this idea to life.
What it does
The SDLC Chatbot is an intelligent assistant designed to help developers and project teams navigate the software development lifecycle. It supports tasks like:
- Problem Analysis
- Brainstorming Solutions
- Code Generation
- Code Refinement
How I built it
The chatbot was built using Google ADK (Agent Development Kit) and Gemini as the underlying LLM. I structured the assistant using sub-agents, each responsible for a specific SDLC phase. Each agent uses prompts and tools tailored to its role, and they communicate with the root agent to maintain context and cohesion across phases.
Challenges we ran into
One of the biggest challenges was that this was my first time using Google ADK. Understanding how agents communicate, how to structure tools, how to integrate sub-agents, and session management involved a lot of trial and error.
What's next for SDLC Chatbot
Integrating the chatbot with FastAPI for a front-end UI
Log in or sign up for Devpost to join the conversation.