Pythagora AI raised $4 million in seed funding for its dev tool that can build full apps
Pythagora AI has raised $4M in seed funding to accelerate the research and development of its AI-powered software development tool which can assist developers by building apps from scratch from ordinary language conversations.
Pythagora is an AI-powered coding partner that can write full apps through ordinary language conversation. It is available as a VS Code extension that harnesses the capabilities of GPT-4 and Pythagora AI's GPT-Pilot. This open-source tool emerged as the team behind Pythagora started researching how much of the app-building process could be reliably automated using the power of LLMs with developer oversight.
Some essential tenets that guide Pythagora AI's approach are that human oversight is necessary, not only to correct mistakes but also because developers may decide to change an implementation after seeing it in action, or may realize that the number of edge cases may require refactoring the implementation before the changes are committed. Additionally, that an AI-powered dev tool cannot figure out a perfect solution does not mean the tool is faulty or ineffective. Most human developers rarely write perfectly working code on their first attempt, thus Pythagora AI takes its approach to AI-powered software development to reflect how developers work, which should be considered a feature, rather than a bug. As a result, Pythagora AI focuses on giving the developers the right tools to harness the power of LLMs rather than spending all of its resources on coaxing near-perfect outputs from the models on a first-attempt basis.
To start coding with Pythagora, developers only need to describe what they want to build in conversation, and the tool will take care of the rest. For instance, for any gaps to be filled after the initial description, Pythagora will ask clarificatory questions before proceeding. Then after writing production-ready, modular code, it will ask the developer before executing the code. Then, Pythagora will review and fix any errors in the code, enabling it to learn from its mistakes. Pythagora AI is working on adding cloud deployment and automated test creation options to the extension.
Pythagora AI's investors include Y Combinator, Inovo, 500 Emerging Europe, Moonfire, Rebel, and Uphonest Capital. The startup plans to use the funding to accelerate research and development to drive growth and user adoption in the short term.
