Many growing organisations have a large inbound candidate pipeline, resulting in numerous CVs coming through without an easy way to understand the candidates behind them. At Lynq, we face the same issue and have decided to use our own tool to sift through this haystack and find the needles.

In this guide, I'll walk you through how to build your own AI-powered recruiting analyst. This tool will automate the process of extracting key information from CVs and conduct an initial web search to build a more comprehensive profile of the candidate in question. If you prefer a video format, please skip to the bottom of the page for an end-to-end walkthrough.

Let's begin with an example CV, using the controversial billionaire, Elon Musk. To summarise the problem we’re addressing: imagine a random applicant named Elon Musk has applied for a job, and we'd like to know a bit more about him.

1. Adding the Google Drive Trigger

First, we need to ingest the CV into the Lynq platform for processing with its AI actions. This can be achieved using the Google Drive trigger, which waits for new files to appear in a Google Drive folder and initiates the workflow whenever it detects a new file. Below, there's a GIF showing how this Google Drive trigger can be implemented.

2. Extracting Key Information

Next, we need to extract all the pertinent information against which we benchmark a CV. For this example, we aim to understand basics such as:

• Candidate Name

• Candidate Email Address

• Alma Mater

• Highest Degree Awarded

• Latest Experience (Company and Role)

• Industries Worked In

You can expand this list to include any additional information you'd like to extract from the CVs. The system can also handle more complex queries that require reasoning, like 'Is this person a good fit for role X, which requires skills Y and Z?'

3. Explode

The 'Extract Keywords' step is designed to return batches of output as it can identify multiple entities (e.g., people) within a single document. For instance, meeting notes might mention several attendees, and this step could extract key details about each participant from a PDF. We use the Extract node to ensure all subsequent steps focus on one item from the list at a time.

4. Profile Enricher to the Rescue 🦸

Here's where Lynq truly shines. By adding a profile enrichment step, the system will automatically conduct various web searches, pinpoint relevant sites about the person, and then compile a summary of its findings. It scans all available content, be it complex news pages, LinkedIn profiles, or personal websites. So, if the candidate has a web presence, you won’t miss out on any crucial insights. Now, let's get it added!

5. Spreadsheets - the destination

With the key information from the CV and a web-enriched dossier in hand, we'll transfer it to a Google Sheet. This way, we begin to compile a searchable spreadsheet with all the essential data extracted from the received CVs. This step is a tad more intricate as we want the spreadsheet to consolidate data from both the 'Extract Keywords' and 'Enrich Person’s Profile' steps. We achieve this by setting them both as inputs for the 'Write to Google Sheet' step.

That completes the workflow, and it didn’t take long. Now you can test it. Click 'Test' to run it against the most recent CV added to your chosen Google Drive and watch the workflow operate in real-time. It intelligently extracts the specified data from the CV, searches for information about the individual online, and then sends the data to your spreadsheet.

If you're satisfied with the tested workflow, you can now publish it using the large 'Publish' button located at the top right of the screen. Once published, the workflow will check every 5 minutes for new files in the Google Drive folder and run the workflow if any fresh CVs are discovered. And there you have it!

Video Walkthrough 📽️

For those interested in a complete end-to-end demonstration of building the workflow, please watch the video below.