What it does

Our project is an AI assistant that simplifies the work of hiring and HR specialists.

In their work, HR specialists are facing with a large amount of data that needs to be systematized and compared with each other.

Often, this data is not well-structured or unstructured at all. For example, CVs that candidates send in PDF files. We looked at several hundred CVs and were amazed at how differently people present seemingly similar information.

The goal of our project is to help process incoming information, resumes and CV by using AI, e.g.

  • extract the necessary information,
  • enter it into the database,
  • make selections from a large collection of resumes,
  • match them with existing vacancies,
  • and determine candidates that are most suitable for a particular vacancy.

The result of the project will be a Web based workplace for HR specialists where they will be able to give tasks to the AI for performing typical routine tasks.

How we built it

Web page user interface Connection to storage for receives CV’s and resumes Integration with Open AI API (ChatGPT-4)

Challenges we ran into

1) Integrarion. Connection of WEB page with CV upload storage, integration with Open AI API (ChatGPT-4).

2) Validation of results how precise of calculations by Open AI it needed manual validation ( manual read of particuar CV's). We did small set of manual validation, more testing need to be done.

Accomplishments that we're proud of

We managed to develop prototype:

1) simple WEB page ( based on mocked data currently):

  • where user can synhronize CV storage and sent to Open AI storage
  • user has possibility to search of skills (multi selectable field)
  • user can specify minimum required job expericence

2) Promt text to Open AI to get needed result - tried several promts to improve the output with different parameters

3) Made integration with Open AI API (ChatGPT-4) (based on mocked data)

4) Pending: continuation of integration of Fronend (User interface) with Backend so that search works based on user entered data.

What we learned

1) we got hands on experience with Open AI, e.g. mass processing of many files as knowledge database

2) prompt engineereing experience

3) pitch deck preparation, market opportunity research

What's next for Project12 (HR assistant)

1) Need to continue working for end-to-end itegration (finalizing prototype)

2) Develop proper integration with different types of storages

3) Develop proper user authorization and authentification in Web portal

4) Try to get first clients

5) Continue with developing more features based on users' feedback

Built With

Share this project:

Updates

posted an update

Day 1: Today we have finalized our idea with the team. We will be focusing on processing unstructured data - submitted CV’s and resumes no find best matches according to criteria defined by HR manager.

Solution meant to be for small and medium business and will be based on pay as you go principle.

We drew architecture of our HR assistent solution and mockup of our web page.

Already splitt responsibilities - web, storage, AI part.

There was first checkpoint with mentor.

The wibe is soo good and pizza was tasty :)

Log in or sign up for Devpost to join the conversation.