Buddhi Kavindra Ranasinghe
Buddhi Kavindra Ranasinghe is an individual AI/ML software engineer, technology evangelist and blogger who builds privacy-first, local/edge AI tools. His stated mission is to make capable AI run on users' own devices, with data remaining local, while educating people about edge AI and open-source tooling.
At a Glance
- Software engineers and Google Antigravity users
- Researchers, academics and students
- Writers and knowledge workers
- Privacy-sensitive users and organizations
- +2 more
AI Tools by Buddhi Kavindra Ranasinghe
(1)
BuddhiAI CLI
Code Graph CLI for AI Agents
Discussions
No discussions yet
Be the first to start a discussion about Buddhi Kavindra Ranasinghe
Latest News
BuddhiAI 2.1.0 introduced dedicated research modes and Text Humanizer Studio.
Introducing BuddhiAI CLI, an open-source code-graph and Google Antigravity spec-driven development harness.
Introducing BuddhiAI Studio, a fast local LiteRT-LM/Gemma inference service for desktops.
Buddhi AI Prompt Builder launched as a free, browser-local Gemma 4 prompt generator.
Products & Services
Browser-based, client-side AI suite initially offering Chat, Summarizer and Writer tools; later supports Gemma 4 E2B, reasoning mode, document upload and local RAG so documents remain in the browser.
Open-source Python CLI that parses repositories with Tree-sitter into a code graph, exposes topology-aware search and compressed reading through an MCP server, generates documentation and project memory, and scaffolds a Google Antigravity agent harness with a spec-driven development lifecycle.
Open-source desktop local-inference service packaged for Docker. It uses LiteRT-LM and Gemma edge models, provides CPU-targeted inference, local embeddings, chat and model management, and OpenAI-compatible endpoints.
Free in-browser prompt generator using a client-side Gemma 4 implementation; intended as a control layer that improves prompts sent to other AI tools without subscriptions or credits.
Market Position
An individual-led, open-source alternative to cloud-dependent AI assistants and coding agents. The differentiator is a local-first/edge approach: browser or desktop inference, privacy-preserving document RAG, no required external API keys for local tools, and developer infrastructure that combines a code graph, MCP and spec-driven agent harness rather than only a chat interface.
Leadership
Founders
Buddhi Kavindra Ranasinghe
AI-focused software engineer, researcher, technology evangelist and blogger from Sri Lanka. His GitHub profile describes him as a Technology Evangelist and Software Engineer (Artificial Intelligence & Machine Learning), and the BuddhiAI history says he began with blogging, founded Buddhilive Developers on July 2, 2015, later worked with TensorFlow, Dialogflow and TensorFlow.js, and is the founder and lead developer of BuddhiAI.
Executive Team
Buddhi Kavindra Ranasinghe
Founder and lead developer, BuddhiAI; Founder and Chief Architect, Buddhilive Academy
Sri Lankan AI/ML software engineer, researcher, technology evangelist and blogger. His public profiles and own history identify him as the creator of the BuddhiAI tools and the author of the BuddhiAI technical blog.
Founding Story
Ranasinghe says his digital work began with Blogger and that he created Buddhilive Developers on July 2, 2015 as a professional home for Google technology updates and tutorials. After becoming interested in Google DeepMind's 2016 work on diabetic-retinopathy diagnosis, he moved toward machine learning and web AI. His later vision was to make AI private, accessible and run where the user is, leading to BuddhiAI's local-first browser and edge-AI projects.
Business Model
Revenue Model
The public materials describe a free, open-source, local-first model: tools run on the user's device, with no cloud lock-in, no credit card and no cost for the ecosystem. The Buddhi AI Prompt Builder explicitly says it has no credits or subscriptions. No paid enterprise or API monetization is stated in the reviewed sources.
Pricing Tiers
Open-source/free tools; the Prompt Builder runs locally in the browser and is described as having no credits or subscriptions.
Target Markets
- Software engineers and Google Antigravity users
- Researchers, academics and students
- Writers and knowledge workers
- Privacy-sensitive users and organizations
- Healthcare and offline-workflow users
- Developers experimenting with local and open-source LLMs
- Private chat and writing assistance in a browser
- Chatting with confidential PDFs and other documents using local RAG
- Academic research, literature analysis and publication-oriented writing
- Generating and improving prompts for cloud or local AI tools
- Running LLM inference locally on desktops without a GPU or external API key
- Offline or privacy-sensitive healthcare and development workflows