Optimizing App Development with Generative AI
summary
Freshworks ships Freddy Copilot for Developers, an IDE-integrated system that turns use-case sentences or Figma/hand-drawn screens into HTML/JS skeletons, OAuth/app parameter templates and runnable code blocks. Engineering uses RAG with a vector DB, Hugging Face SentenceTransformers MiniLM L6, and multiple managed LLMs to keep docs and API schemas current. It runs in production, is in public beta for Freshworks web apps, has produced 225 apps so far and claims 3x productivity for developers.
AI-written summary · about
chapters (7)
- 0:00Introduction to GenAI at Freshworks
- 3:00Why Freddy Copilot was built
- 9:17Engineering challenges and solutions
- 18:05Building the knowledge base with RAG
- 22:50Auto-generating app templates
- 26:47Results and productivity gains
- 30:38Q&A and closing
related talks
12mGenAI - Web Development using Freddy Copilot
21mRevolutionizing software creation: Harnessing the power of GenAI
17mAI-Driven Frontend Development: Beyond Traditional
11mReshaping the Developer's Role: Adapting to No-Code & AI
34mUsing Appsmith for building GenerativeAI apps
57mOne Team’s Journey to Using AI Without the Hype
25mThe Ai-Assisted Developer Workflow: Build Faster and Smarter Today
24mSupercharging Dev Workflows: AutoGPT Meets React Flow
