3 ms·
Why would you come to that conclusion? Its really not that hard to understand. They already had a library of templates, modules, components, and existing code,
by pyman 1y ago
Why would you come to that conclusion? Its really not that hard to understand.
They already had a library of templates, modules, components, and existing code, which they were likely reusing.
According to their website, their virtual assistant interacts with customers to understand requirements, recommends suitable templates, and assembles a basic version of the app automatically. Then assigns developers to that project, as shown on their site. In one of their blog posts, they mentioned using machine learning to generate dependency graphs, which was used to map out what needed to be built and estimate timelines.
Since 2019, they've expanded their AI and ML team and hired the former Director of AI from Amazon.
The issue here is that you're quoting articles published before GenAI or ChatGPT even existed. Back then, AI mostly referred to machine learning, it was a different landscape entirely.
I'm only interested in understanding why they lied about their finances. Because the tech they had was actually quite impressive.
We'll never know if they were reusing 40, 60, or 80 percent of the code. What we do know is that developers were spending 7 months (based on a screenshot from their website) writing code. So GenAI clearly wasn't used to generate the code. But to be fair, they never claimed it was.
- YeGoblynQueenne 1y agoSo you mean that Builder.ai is using more AI now than they used in 2019? In other words, more than "roughly 82%" of apps the company now develops is built "autonomously, in the first hour by Engineer.ai’s technology"? Note that generative AI and LLMS like GPT, Elmo and BERT existed earlier than 2019 and were the subject of much research (see "BERTology").
- pyman 1y agoI'm not sure what you're referring to. I don't think you understand the difference between automation, traditional AI, and generative AI. Back in 2019, when someone said part of an app was built "autonomously", it usually meant they reused components and generated the glue logic, configurations, or some custom code around it. In BuilderAI's case, they said they were using AI/ML to create dependency graphs. Just read their blog, it's all there! The Transformer paper came out in 2017, and the first time I heard of GPT was in 2019. At that point, only a few companies like Google and OpenAI were working on LLMs. If you're expecting BuilderAi, a small startup from India with limited funding, to compete with multibillion dollar companies in 2019, then you're being delusional. I don’t think I can be of any further help. Apologies.