Colin Buresh is the Head of Talent and HR at DeepScribe. Buresh started out as a casualty adjuster and a sales rep before becoming a recruiter. He has spent a little over a decade in recruiting; this includes co-founding a recruitment agency and a Lead Technical Recruiter position with Expanse (Acquired by Palo Alto Networks).
How can Colin do it all and still retain a semblance of normalcy?
Whether you call it People Operations or HR, the team responsible for managing an organization’s workforce and ensuring the well-being of its employees is considered to be an expensive function by most businesses. The reason? It’s not directly tied to revenue.
In his experience, the more you can do to enable and streamline HR, the less you have to spend on it for good results. For example, running the kind of de-biasing experiments Buresh is running right now would have cost a pretty penny for a company of DeepScribe’s size a few years ago. Only large companies the size of Alphabet had access to resources and tools to perform such experiments. Generative AI might still be firmly in the emerging technology camp, but it’s leveling the playing field for strapped teams like Buresh’s.
When Buresh dreams of GenAI, this is where he sees it making the most impact:
- Talent attraction
- Employee support
He’s already made tremendous strides in incorporating ChatGPT as a tool in talent attraction, but during our conversation, he also sketched out his employee support and retention plans. We wrote up a summary of our conversation along with Buresh’s advice on incorporating AI into HR processes.
Weeks to hours: Talent attraction with ChatGPT
This is how most TA teams handle recruitment: A recruiter works with a stakeholder to create an interview process and a target profile. The recruiter starts outreach. Candidates flow down the interview process till the hiring manager finds a skilled candidate whose expectations are in line with the company’s offer. Interviewing candidates of course takes time. But there is a hidden time investment that Buresh feels is commonly overlooked – the process of creating a job description and interview process. There is a lot of back and forth between the recruiter and the hiring manager. Between getting a job description, interview process, and an interview team ready to start a search, those processes can sometimes that can take weeks. Even worse, they can be thrown together or have unconscious bias built in throughout the process.
It’s not that the process is different each time.
“It’s usually going to involve a recruiter phone screen, a hiring manager phone screen, a work or product interview, and a final panel interview with some executives. These are steps every candidate is going to go through, and the hiring manager knows that. But big companies usually have a process and lots of best practices in place. At a smaller company like DeepScribe, it’s probably the first time that somebody has gone through this process. It might even be the first time the hiring manager has ever even hired. So how does the recruiter help the hiring manager create any effective process when there’s no history of best practices? This is where ChatGPT comes in.”
At DeepScribe, the right prompts helped the TA team take the interview creation process from a couple of weeks down to hours. Not days. Hours.
Another area that GenAI pitches in is in reducing bias in the wording of job descriptions. The team uses Gender Decoder and ChatGPT to figure out how to make it appeal to people from a wider background.
“Just asking it how to reduce bias gets us some interesting and helpful feedback.”
Employee support and retention
Employee support, Buresh admits, is not an AI focus area right now for DeepScribe. But he has big plans.
“Employees want easy access to content and meaningful training, which is very much in line with what companies want. Every company wants to be able to offer more to their employees so they can do the best work of their lives and gain new and valuable skills. There are a ton of programs out there, but it’s only a small portion of the workforce that gets excited and accesses it. How many people do you know have actually finished a certification? I think GenAI will have a huge impact on people development and making information easily available to employees.”
Employee surveys are also key stomping grounds for GenAI in Buresh’s mind. ChatGPT makes generating and reviewing key surveys like mid-year reviews a much easier exercise than it used to be.
GenAI can really boost performance reviews. Imagine an employee who wants to punch up their talking points before they submit them to their manager or HRBP. Before ChatGPT, they might have had to depend on the kindness of their friends to help them out. Now, it’s only a URL away.
AI can have a huge impact, in Buresh’s opinion, on the way performance reviews are run. Can we peer in to find out if the manager who gave a bad review and fired the employee (many a Reddit post starts here) was biased against the employee or had just cause and is a poor communicator? Sometimes, employees don’t have the transparency they really need from their managers.
Buresh is hesitant to quantify ChatGPT’s impact.
“Smaller companies now have access to the kind of tools only large companies have had for free. Isn’t that enough? But if we had to zoom out, AI tools (probably not ChatGPT) will impact all the big soft-skills-influenced scores like regretted attrition or employee satisfaction scores and key D&I metrics.
We’re still in the early stages of AI. But ChatGPT is a great start, and the more we integrate it into our processes, the faster those future steps that make workplaces fair and equitable will appear in new tools.”