
ChatGPT agent mode: The “little helper” that transformed recruiting & solved parking nightmares
Claire Vo (host), Michal Peled (guest), Claire Vo (host)
In this episode of How I AI, featuring Claire Vo and Michal Peled, ChatGPT agent mode: The “little helper” that transformed recruiting & solved parking nightmares explores chatGPT agent mode automates recruiting, personas, and parking planning workflows The episode introduces ChatGPT “agent mode” as a practical way to offload repetitive, high-friction work by letting ChatGPT browse sites, log in, and take actions with human handoff points.
ChatGPT agent mode automates recruiting, personas, and parking planning workflows
The episode introduces ChatGPT “agent mode” as a practical way to offload repetitive, high-friction work by letting ChatGPT browse sites, log in, and take actions with human handoff points.
Michal shows a recruiting automation where an agent logs into LinkedIn, searches against an uploaded job description, applies hiring-team constraints, and returns a ranked shortlist of candidates in minutes.
Next, he turns expensive customer research documents into five interactive buyer-persona GPTs, using NotebookLM for source-grounded prompt drafting and then adding guardrails to keep personas accurate and safe.
Finally, he uses ChatGPT to generate a filtered baseball-game schedule and output an ICS calendar that warns employees about Oracle Park game days that trigger parking price spikes.
Key Takeaways
Treat agent mode like a coworker with a clear job description.
Michal frames prompts as delegating to a “little helper” (e. ...
Get the full analysis with uListen AI
Add explicit handoff points for secure, collaborative automation.
For sensitive steps like logging into LinkedIn, the prompt instructs the agent to pause and let the user take control, keeping the flow practical while still automated end-to-end.
Get the full analysis with uListen AI
Restrictions beat general instructions for repeatable sourcing quality.
Candidate constraints (location, recent activity, seniority, tenure/unemployment windows) were taken from the hiring team’s real process, making the agent’s search behavior match how the team actually screens.
Get the full analysis with uListen AI
A “match score” makes AI outputs more usable, even if imperfect.
Requesting a percent match helps recruiters quickly compare candidates and decide where to inspect deeper; it’s not exact science, but it’s a helpful prioritization layer.
Get the full analysis with uListen AI
Persona GPTs work best when the persona lives in instructions, not PDFs.
Michal found that simply uploading research leads to answers *about* the persona; tightly written instructions (“you are this person…belief system…decision style…tone”) produces answers *as* the persona.
Get the full analysis with uListen AI
Use NotebookLM to prevent hallucinations during research-to-prompt translation.
NotebookLM’s source-only answering and citations let him verify each persona trait is grounded in the original research and avoid the model “filling in gaps” creatively.
Get the full analysis with uListen AI
Guardrails are essential when teams will ‘mess with’ personas.
He adds boundaries (no political/religious/gender/racial commentary, respectful tone, no slang, not a general assistant) because colleagues will stress-test personas with off-topic or edgy prompts.
Get the full analysis with uListen AI
Notable Quotes
“I want a little helper. I'm a recruiter. I want someone who is like me.”
— Michal Peled
“If you can codify what a person's step-by-step workflow is… you can replicate and automate that at scale.”
— Claire Vo
“Don't add or modify text that is not written or implied in the text. Okay, I know you're creative. I'm turning you down.”
— Michal Peled
“Out of these five, four of them were never found by us manually… and they really fit the description.”
— Michal Peled
“It can ruin your entire day, for sure.”
— Michal Peled
Questions Answered in This Episode
In your recruiting agent prompt, how did you operationalize “70% match” so it didn’t become arbitrary—and what would you change to make scoring more explainable?
The episode introduces ChatGPT “agent mode” as a practical way to offload repetitive, high-friction work by letting ChatGPT browse sites, log in, and take actions with human handoff points.
Get the full analysis with uListen AI
What are the key failure modes you saw when the agent searched LinkedIn (e.g., getting stuck, wrong filters, partial profile access), and what prompt guardrails reduced them most?
Michal shows a recruiting automation where an agent logs into LinkedIn, searches against an uploaded job description, applies hiring-team constraints, and returns a ranked shortlist of candidates in minutes.
Get the full analysis with uListen AI
How do you handle compliance and privacy concerns when an agent logs into LinkedIn and navigates candidate profiles using a user’s account?
Next, he turns expensive customer research documents into five interactive buyer-persona GPTs, using NotebookLM for source-grounded prompt drafting and then adding guardrails to keep personas accurate and safe.
Get the full analysis with uListen AI
For the persona GPTs, what specific instruction patterns most improved “talking as the persona” versus “talking about the persona”? Can you share a before/after snippet?
Finally, he uses ChatGPT to generate a filtered baseball-game schedule and output an ICS calendar that warns employees about Oracle Park game days that trigger parking price spikes.
Get the full analysis with uListen AI
When NotebookLM produced persona prompts longer than ChatGPT’s 8,000-character limit, what did you cut first without losing fidelity—tone, journey map, tech stack, or decision rules?
Get the full analysis with uListen AI
Transcript Preview
We're gonna start with something that we haven't actually seen on How I AI yet, which is agent mode in ChatGPT.
My use case was with our hiring team. Part of their workflow is to browse through many LinkedIn profile and search for relevant candidates. It takes a lot of time.
Let's talk about the prompt. I'd love for you to go through how you thought about structuring it to make it effective with the agent.
I want a little helper. I'm a recruiter. I want someone who is like me. So I started by telling it, "You're an IT recruiter," and then I described what I want it to do.
I love that you called it your little helper, because don't we all want an AI little helper? [upbeat music] Welcome back to How I AI. I'm Claire Vo, product leader and AI obsessive, here on a mission to help you build better with these new tools. Today, I have Michal Peled from HoneyBook, their Technical Operations Engineer, who's building tons of internal tools and automations to make their team's life easier and reduce friction. Michal's gonna show us some advanced features of ChatGPT, including agent mode, replicate not one, but five of their personas as AI identities, and save me a lot of time on my commute using ChatGPT. I'm really excited about this episode. Let's get to it. This episode is brought to you by Brex. If you're listening to this show, you already know AI is changing how we work in real, practical ways. Brex is bringing that same power to finance. Brex is the intelligent finance platform built for founders. With autonomous agents running in the background, your finance stack basically runs itself. Cards are issued, expenses are filed, and fraud is stopped in real time without you having to think about it. Add Brex's banking solution with a high-yield treasury account, and you've got a system that helps you spend smarter, move faster, and scale with confidence. One in three startups in the US already runs on Brex. You can, too, at brex.com/howiai. Michal, thank you so much for joining How I AI. I'm excited to see what you have to share.
Thank you so much for having me.
We're gonna start with something that we haven't actually seen on How I AI yet, which is agent mode in ChatGPT. And so I'm wondering if you can just go ahead and dive into what was the problem that you were trying to solve, and why was this agent mode, this agentic browsing, the solution to the problem you were having?
Our problem was, um, you know, same as, same as our customers are having. Uh, you have to do your job. You have a job that you really love doing, and you have your proficiencies and, uh, and expertise. However, you spend a lot of your time doing the, the, um, mundane, thoughtless, uh, manual, repeating work in order to do... uh, to get the information that you need. So my use case was with our hiring team, and as a recruiter, when you get a job description that you need to recruit to find candidates for, part of their, part of their, uh, workflow is to, uh, browse through, uh, many LinkedIn profile and search for, uh, relevant candidates that may be, uh, relevant for the job descriptions, and it takes a lot of time. It can be hours of browsing through profiles and going through all of the characteristics that they're looking for. So I wanted to take that load off of them, and, uh, ChatGPT agent mode came just in time. We all talk about what agent is and what agents do and how we can use them. In ChatGPT, it's very, uh, simple to understand. So you just open a chat with ChatGPT, but then you add an instruction and turn it into an agent mode very simply from the toolbar. And once it goes into agent mode, it means that it can take the prompt, or you can actually use specific prompts, to tell it not just to search for information online, but also to perform actions for you. And why did I need it in this case? Because I needed to log in to LinkedIn. I don't want it to just search for profiles, uh, on LinkedIn, just, just profiles that are publicly accessible. That's not the information that I need. So I needed it to log in into LinkedIn, and I needed it to perform search, and I needed it to go through the profiles and, and look for the restrictions that I wanna give it, and those restrictions were provided by the actual, uh, hiring team, that they actually use it as, uh, requirements for potential candidates that they find.
Install uListen to search the full transcript and get AI-powered insights
Get Full TranscriptGet more from every podcast
AI summaries, searchable transcripts, and fact-checking. Free forever.
Add to Chrome