These days, businesses are using AI for many tasks, but most of them don't know how to use it to improve productivity and quality of work. They're not fully aware of AI's capabilities and how to maximize its potential.

We've been thinking about how to integrate AI into our HRMS software. Let me explain what HRMS is.

HRMS is a SaaS resource management software that helps companies manage their employees and recruitment process. It has many features that can boost your company's productivity.

We're in the midst of the AI era, and it's clear that businesses aren't taking full advantage of AI.

The issue we encountered during our recruitment process was...

We noticed that some processes in our resource management software were taking up too much time and effort. For example, when candidates were sending us their CVs, recruiters had to manually fill out all their details, such as name, picture, education, contact number, email, past employment history, skills, current salary, expected salary, and more.

Manually filling out all this information was becoming a real hassle, so we started looking for a solution to make the process more efficient by incorporating AI.

After brainstorming, we finally identified a solution: using AI to streamline the process. We worked on integrating this solution into our software, and now we use it on a daily basis. It has saved us a lot of time and money.

Now, let's walk through the process of how we implemented this solution in our software.

Extracting the information from the CV

We receive most CVs in PDF format, so we have to convert them to computer-readable format in order to extract the necessary information. To do this, we use LangChain, a framework that makes it easier to create applications using large language models. LangChain is great for tasks like document analysis, chatbots, and code analysis.

Thanks to LangChain, we were able to analyze the documents and find all the information we needed.

We used Vector Embedding…

Yes, you read it correctly! We utilized Vector embedding with the help of the ChatGPT API. Vector embedding is a technique used in text analysis to represent words as real-valued vectors. These vectors encode the meaning of the word in a way that words closer together in the vector space are likely to have similar meanings.

Take a look at the image below to see how it works. All you have to do is upload the candidate's CV, and it will extract all the information and fill out the form in just seconds. You can see the extracted information filled out in the form in the right section.

It accurately identifies the information and adds it to the respective section.

We effectively utilized the Prompt Engineering and Retrieval-Augmented Generation (RAG)

We utilized prompt engineering to guide AI in extracting the necessary information in the exact format needed for easier analysis by recruiters.

Retrieval-augmented generation (RAG) combines an information retrieval component with a text generator model to tackle knowledge-intensive tasks.

By using both of these methods effectively, we were able to achieve the desired output. The image below shows that after retrieving the information, a candidate profile is created with all the necessary details. This profile can be easily accessed by recruiters in the future.

Before implementing this solution, it used to take 15-20 minutes to save all candidate information. After implementing this solution, it now only takes 8-10 seconds to complete the process. Just think about how much money and time this can save for your business!

Human resource managers can leverage artificial intelligence to make proactive and well-informed decisions regarding talent management and business growth. AI has the potential to improve the HR representative experience, ultimately resulting in better outcomes for companies. By streamlining the candidate selection process, companies can enhance their employee recruitment efforts in a more efficient manner.

Have you ever worked in HR with AI? Share your thoughts and experiences below! Thank you for reading!