Posted by 
For the last 8 years, we’ve been helping companies hire developers based on their potential and skill rather than “pedigree” on a resume. We’ve been able to do this and build a multi-million dollar business (without VC money) because we went beyond the obvious.
We built intelligence into our products so that we could spot potential in a candidate. We have product features that read beyond what a resume just says, and we have Artificial Intelligence that can parse a developer’s code and identify their true potential and skill. It’s taken us years of manual code analysis, trying out different ML models, building out AST parsers for every language, and finally we were able to predict a developer’s ability to write good, clean code. We’ve evaluated tens of millions of lines of code and we know how to assess for real world coding skill.
But we had limitations. We didn’t have the resources to uncover all aspects of a developer candidate. We could not go deep into understanding their potential, and we could not go wide to cover different technology stacks. We were limited by the resources we had to build out the tech. Untill now.
The recent AI advances with GPT (and the likes) will allow us to use our intelligence, that we’ve honed over 8 years, and apply it across a spectrum of use cases that would otherwise have taken us years to cover, if at all.
For example – we have ML models that can predict whether you should interview a candidate based on any code they have written. However, we support only 6 coding languages. Now, with OpenAI APIs we can support any language. This is but a simple example.
However, we have to go beyond the simple.
And this is the crux – if we’re building with AI, we have to imagine a new future and build for that. Putting a chat bot into existing systems is not the future. Providing our users super personal, extremely functional experiences that they would not have imagined possible is the future.
What is the biggest problem in tech hiring that we can solve with AI’s help? We’re figuring this out. We’re also doing deep work with prompt engineering, tons of user research, sketching out protoypes and letting our imagination fly. Our MVP will be out in 6 weeks that showcases our vision for the future.
As we lead up to our launch, I’ll be writing more on our learnings using GPT, the direction we’re taking, the specific problems we are looking to solve, our solution and so on. If you’d like to know more about what we’re building, or want to be part of the private beta, ping me on Linkedin or email me at hello@geektrust.in, and I’ll add you to our mailing list.