Proof AI

• 

  Generated simulation

• 

  Hire score

• 

  View candidates

• 

  Insert description

• 

  Benefits

• 

  Feature list

• 

  Hero section

Inspiration

Proof AI was inspired by the real trouble that comes with the hiring process that has been put aside for too long. Being both a student and employee myself, I have experienced the onboarding and recruiting processes of both educational and corporate organizations. Companies spend hundreds of thousands of dollars and hundreds of hours of wasted time on faulty hires, topping the list for the most impactful inefficiencies within a company. By implementing AI, we can now assist companies in the recruiting process by proving competency of candidates before they even enter the interview room.

What I Learned

Contrary to how a lot of people view "one and done" full stack app dev tools, I learned a lot. Writing code is a skill, but the best way to learn is by communicating with someone or something. In this case, I can talk to the AI agent as if it is my own personal engineer, which helps me understand the software by allowing me to view the processes that it undertakes in plain English, not code gibberish.

How it was Built

300 credits is incredibly generous, although I tend to rip through credits like crazy since I converse with the agent on any little feature. Although by using the allotted credits, I was able to first craft the landing page and the general feature list, and then fine tune the separate pages of the application for their purposes. By using this flow, I was able to first tackle the roots of the application, then as you move upwards you focus on the fine tuning of the different branches to refine the user experience. I call this the "tree" method, I find it the most efficient way to lead a development projects.

Challenges

The biggest challenges that I had faced was making sure that the main function; generating simulations, actually worked. MeDo handled this process incredibly and when prompted, it underwent troubleshooting to fix the issue. The most difficult part was the implementation of AI to generate the simulation. Several models didn't work, making the process research heavy and time consuming as I had to do a lot of work to make sure that MeDo and the API were in collaboration.

Built With

• groq
• medo