Using AI in Software Design: How ChatGPT Can Help With Creating a Solution Architecture
summary
Shows a working process that uses ChatGPT (GPT‑4 via API) plus embeddings and a vector DB to convert an ISAQB advanced exam PDF into a solution in about two days of iteration, producing quality scenarios, a utility tree, PlantUML diagrams and an AsciiDoc draft. Model reads dropped images, asked 10 clarifying questions, outputs downloadable PlantUML files and code interpreter results, but hits context limits (32k→128k tokens via API) and requires careful priming, iterative prompts and human validation.
AI-written summary · about
chapters (4)
- 0:00Intro
- 0:30Session "Using AI in Software Design: How ChatGPT Can Help With Creating a Solution Architecture" by Ralf Müller
- 42:09Ralf answers questions from the audience
- 47:17Outro
related talks
46mUsing AI in Software Design: How ChatGPT Can Help With Creating a Solution Arch.
53mBuilding an App with ChatGPT
01h 00mNext-Gen Software Architecture with Large Language Models (LLMs) like ChatGPT
32mHow to Make AI Work for You: A Deep Dive Guide into ChatGPT for IT Professionals
15mHarnessing the Power of Generative AI for Intelligent Applications
36mAI for Government Efficiency with ChatGPT: From Hype to Reality
51mA ChatGPT on your own data: how does it work?
01h 09mChatGPT and OpenAI API solutions: 👑 My Favourite Implementations