Hey everyone, welcome back to My Weird Prompts. I am Corn, and I am joined as always by my brother, the man who probably has more browser tabs open than most small businesses have files. Seriously, Herman, I saw your screen earlier and I think I saw a tab from two thousand twenty-two that you are still ‘saving for later.’
Herman Poppleberry here, and you are not wrong, Corn. I am currently sitting at eighty-seven tabs, and at least forty of them are deep-dive research papers on agentic workflows and labor market shifts from the last forty-eight hours. It is not clutter, Corn; it is a high-velocity information ecosystem. It is a lifestyle choice.
It is a choice that makes my cooling fans spin just looking at it. Well, we have a really timely one today. Our housemate Daniel sent us a voice note about something that I think is weighing on a lot of people right now, especially as we settle into February of twenty twenty-six. He is looking at the job market, specifically the remote landscape, and asking how we can use AI to flip the script on the hiring process.
I love this angle. Usually, the narrative is about how companies use AI to screen candidates—the dreaded automated tracking systems, the AI-led video interviews, the sentiment analysis on resumes. It can feel a bit like being a bug under a microscope. But Daniel is asking how the bugs can get their own microscopes. How do we use these tools to vet the companies before we ever sign that contract? How do we perform a ‘reverse background check’?
Exactly. It is about closing that information gap. When you are applying for a remote role in twenty twenty-six, you do not get to walk through the office and see if people look miserable or if the coffee machine has been broken for three weeks. You are sitting in your own house, looking at a polished website and a recruiter’s ring light. Daniel mentioned evaluating things like employee retention, funding history, and that elusive thing called company culture. He wants to know if the ‘vibe’ matches the ‘math.’
It is the ultimate due diligence challenge. And you know, we have touched on some of the broader shifts in the labor market before. Back in episode three hundred ninety-seven, we talked about the ‘Great Hollowing’ and how AI is changing the career ladder itself. But today is much more tactical. It is about the hunt. It is about using LLMs and autonomous agents to peel back the corporate wallpaper.
Right. So let us start with one of the biggest signals Daniel brought up: employee retention. Traditionally, you might look at a company's LinkedIn page and see they have five hundred employees. But that does not tell you the velocity of those employees. Herman, how can AI help us get a more granular view of that in today’s environment?
This is where a good AI agent—something with real-time web access like a Perplexity Pro or a custom Claude-based scraper—really earns its keep. If you just ask a standard chatbot, ‘what is the retention at Company X,’ it might give you a vague answer based on a two-year-old training set. But if you use an agentic workflow, you can do some fascinating math. You can ask the AI to analyze the tenure of the current staff listed on public professional networks and, more importantly, cross-reference that with the ‘past experience’ sections of people who recently left.
But wait, a company might have a high turnover in sales—which is common—but very stable engineering teams. Does the AI catch that nuance, or does it just give you a scary-looking average?
That is the key. You have to prompt it for departmental segmentation. You can tell the AI, ‘Analyze the last fifty departures from this company. Categorize them by department. What is the average tenure in the engineering department versus the marketing department?’ If you see that senior engineers are leaving every fourteen months, but the job description says they are looking for someone to ‘build a five-year architectural roadmap,’ that is a massive red flag. The AI can process those hundreds of profiles in seconds, whereas a human would take days to click through all of them and probably lose their mind in the process.
That is a great point. And it connects back to what we discussed in episode one hundred, where we talked about AI as a mirror for our own philosophical identities. If you know you are the kind of person who needs stability and deep work, seeing a high-velocity churn through an AI analysis tells you that this mirror is showing you a reflection you might not like. You are seeing the ‘burn and turn’ reality behind the ‘we are a family’ slogan.
Precisely. And Daniel also mentioned funding history. This is huge for remote startups in twenty twenty-six. We are in a weird economic cycle where ‘zombie startups’ are still a thing—companies that have enough cash to stay alive but not enough to actually grow. You want to know if the company is actually healthy or if they are just hiring a bunch of people to make themselves look attractive for a fire sale or a merger.
Right, because a remote worker is often the first to go if the runway gets short. It is easier to lay off someone you have never met in person. It is just a ‘deactivate’ button on Slack. So, how do we use AI to parse the financial jargon and see the actual cliff?
You can feed an AI the recent news cycles and funding announcements from places like Crunchbase or even SEC filings if they are public. But the real magic is asking the AI to calculate the ‘implied burn rate.’ You can say, ‘Based on their series B of twenty-five million dollars in late twenty twenty-four and their current headcount growth from one hundred to two hundred and fifty, what is their likely monthly burn?’ The AI can give you a rough estimate. It might say, ‘They probably have seven months of cash left if they do not raise a Series C by June.’ That is a conversation you need to have with the hiring manager.
That feels like insider information, but it is all based on public data. It is just that most humans do not want to do the math or do not know how to connect the headcount to the burn rate.
Exactly. It is about synthesizing disparate data points. Most people see a headline that says, ‘Company X raises thirty million,’ and they think, ‘Great, they are rich!’ But the AI can remind you that thirty million for a company that just doubled its staff to four hundred people and has a high-cost cloud infrastructure actually goes very fast. The AI acts as your financial advisor, telling you if you are boarding a rocket ship or a sinking raft.
I want to move to the culture piece, because Daniel mentioned something really specific in his prompt. He talked about companies being litigious or manipulating Glassdoor reviews. In twenty twenty-six, we know that companies are using AI to write the reviews now. That seems like a harder thing for an AI to sniff out than just counting months of employment.
It is harder, but it is also where AI is getting surprisingly good at ‘detecting its own.’ It comes down to pattern recognition in language. Think about Glassdoor reviews. We all know companies ‘salt the mine.’ They tell their HR team or use a bot to go write a bunch of five-star reviews to bury the one-star review that mentions the CEO’s temper.
Oh, I have definitely seen those. They all sound the same. ‘Great culture, fast-paced environment, free snacks’—even though it is a remote job and there are no snacks!
Exactly! And that is the pattern. You can take the last fifty reviews of a company and feed them into an AI like Claude or GPT. You ask it to identify clusters of similar phrasing or ‘synthetic-sounding’ sentiment. If thirty of the five-star reviews use the exact same three adjectives or follow the same structural flow, the AI can flag those as likely coerced or fake. Then, you tell the AI to ignore those and perform a sentiment analysis only on the reviews that contain specific, detailed criticisms. You are looking for the ‘unfiltered signal.’
So it filters out the noise to find the truth. That is brilliant. What about the litigious part? Daniel seemed worried about companies that use legal threats to silence former employees. That is a terrifying prospect for a remote worker who might not have the resources for a legal battle.
That is a dark side of corporate culture that often stays hidden. But an AI can search public court records, Pacer, or news databases for the company name alongside keywords like ‘non-disclosure agreement,’ ‘breach of contract,’ ‘non-compete,’ or ‘wrongful termination.’ If a small startup has twelve lawsuits against former employees for supposedly stealing trade secrets or violating non-competes, the AI will find that pattern. A human might only find one or two and think they are isolated incidents. The AI sees the trend of ‘legal bullying.’
It is like having a private investigator who can read ten thousand pages a minute. But I wonder, Herman, is there a risk of confirmation bias here? If I go looking for red flags, the AI will probably find them because every company has some flaws. No company is a perfect utopia.
That is a very incisive question, Corn. And you are right. If you tell an AI, ‘Find me reasons why I should not work here,’ it will find them. To be objective, you have to ask for a balanced SWOT analysis—Strengths, Weaknesses, Opportunities, and Threats. You want to see the retention data alongside the growth data. Maybe the turnover is high because they are a ‘stepping stone’ company. If the AI sees that everyone who leaves Company A ends up at a top-tier firm like OpenAI or NVIDIA, that tells a very different story than if they all end up unemployed. That is actually a positive signal for your career growth, even if the retention is low.
That is a great distinction. It is about the context of the data. It is the difference between a ‘toxic environment’ and a ‘high-performance launchpad.’
Right. And for remote workers, there is one more signal that I think is critical, and AI is the only way to track it effectively: the digital footprint of the leadership. In the remote world, the CEO’s personality is the culture.
You mean like what the CEO is posting on X at three in the morning when they are stressed about a board meeting?
Exactly. But also what they say in long-form interviews, their blog posts, and their responses on professional forums. You can ask an AI to summarize the ‘leadership philosophy’ of a CEO based on their last five years of public statements. Does the CEO talk about ‘autonomy,’ ‘outcomes,’ and ‘trust,’ or do they talk about ‘visibility,’ ‘accountability,’ and ‘productivity metrics’? For a remote worker, that is the difference between a dream job and a micromanaged nightmare where you have to move your mouse every thirty seconds.
I can see that. If the AI flags that the CEO constantly uses words like ‘visibility’ in a way that suggests distrust, you might find yourself with ‘bossware’ tracking software on your laptop within a week of starting. The AI can warn you before you ever open that laptop.
It is all there in the language. Humans are often swayed by a charismatic interview—we want to believe the best. But the AI is looking at the cold, hard history of their words. It does not get charmed by a nice smile or a shared hobby.
So, let us get practical for a second. If someone is listening to this and they are about to have a second interview with a remote-first company tomorrow, what is the actual workflow? Do they just open a chat window and start typing?
I would suggest a three-step process. Step one: The Data Gathering. Use an AI tool with live web access. Ask it: ‘Find the funding history, current headcount growth over the last twenty-four months, and average employee tenure for Company X from public profiles. Provide a departmental breakdown if possible.’
Okay, step one is the numbers. What is step two?
Step two: The Sentiment Audit. Feed it the text of the last thirty Glassdoor or Indeed reviews. Ask it: ‘Identify recurring themes in the negative reviews. Check the positive reviews for linguistic patterns that suggest they were written by the same person or an AI. What is the most common reason people give for leaving?’
And step three? The final check?
Step three: The Leadership Alignment. Ask the AI: ‘Summarize the top three values demonstrated by the CEO in their public communications over the last two years. Compare these to the company’s stated mission on their website. Are there any significant contradictions?’ If the website says ‘we value work-life balance’ but the CEO has three articles about why ‘ninety-hour weeks are the only way to win,’ you have your answer.
That is a powerful toolkit. It almost feels like we are entering an era of radical transparency, whether the companies like it or not. The ‘information asymmetry’ that has favored employers for a hundred years is finally starting to level out.
It is about time, honestly. For decades, companies have had all the data. They have your resume, your background check, your credit score, your social media history. It is only fair that we use the same technology to see who they really are when the lights are off and the recruiters aren't looking.
You know, it makes me think about Daniel's point about not going overboard. There is a fine line between due diligence and paranoia. You don't want to talk yourself out of a great opportunity because the AI found one disgruntled employee from five years ago.
Totally. You are not looking for a perfect company because those do not exist. You are looking for a compatible company. If the AI tells you the company is chaotic but high-growth, and you love chaos and growth, then that is a green flag for you, even if it is a red flag for someone else. The AI isn't making the decision for you; it's just giving you better ingredients for your own judgment.
That is a great way to put it. It is about alignment. Herman, did you see that recent study from January about remote work and AI-driven monitoring? It was saying that companies are increasingly using AI to track employee sentiment in real-time by analyzing Slack messages.
I did! It is a bit Orwellian, isn't it? They are looking for ‘vibe shifts’ in the Slack channels to see if morale is dropping before people start quitting. They are using AI to predict who is going to leave.
Right, so if they are using AI to predict our behavior, using AI to analyze their public behavior is really just balancing the scales. It is like a digital cold war, but hopefully, one that leads to better matches and less ‘quiet quitting’ later on.
I think it will. Imagine a world where the AI can tell you, ‘Hey, based on your working style and this company's actual internal data, you have an eighty-five percent chance of being happy there for at least two years.’ That would save everyone so much time and heartache. We might actually see the end of the ‘bad hire.’
It really would. And I think that is the ultimate goal of what Daniel was asking. It is not about catching them in a lie, necessarily. It is about finding the truth of the relationship before you get married to the job. It’s about entering the contract with your eyes wide open.
Exactly. And speaking of relationships, if you are listening to this and you have a relationship with our show, we would love to hear from you. We are always looking for ways to make this more useful for you. If you have a ‘weird prompt’ that helped you land a job or avoid a bad one, send it our way.
Yeah, and if you have a second, a quick review on your podcast app or a rating on Spotify really goes a long way. It helps other people find us and helps us keep doing this deep dive into the weird and wonderful world of AI prompts. We are trying to hit our goal of five thousand reviews by the end of the quarter, and we are so close!
We see every single review, and we appreciate them more than you know. Even the ones that tell me to close my browser tabs.
Especially those! So, to recap for Daniel and everyone else out there in the job hunt: use the AI to do the math on retention, use it to parse the financial runway, and use it to spot the linguistic patterns in reviews and leadership talk. The information is out there; you just need the right tool to synthesize it. Don't just read the job description; audit the company.
And do not be afraid to ask the AI the hard questions. Ask it, ‘Why might a talented engineer hate working here?’ Or, ‘What is the biggest risk to this company's survival in the next twenty-four months?’ The answers might surprise you, and they will definitely give you better questions to ask in your next interview.
This has been a fascinating look at the other side of the hiring coin. Herman, thank you for diving into the weeds on those technical signals. I know you have about eighty-seven tabs to get back to.
Only eighty-three now! I closed four while we were talking about burn rates. Progress! I am practically a minimalist now.
That is a win in my book. Well, thanks everyone for listening. You can find all our past episodes, including the ones we mentioned today, at myweirdprompts.com. We have a full archive there, and you can even send us your own prompts through the contact form.
We are also on Spotify and most other podcast platforms. Just search for My Weird Prompts. We drop new episodes every Tuesday and Friday.
Until next time, keep asking the weird questions.
And keep looking for the signals in the noise.
This has been My Weird Prompts. See you in the next one!
Bye everyone!
I really think the point about the CEO's three a.m. posts is going to be the most used tip from this episode, Herman. People love a bit of digital detective work.
It is the most honest data point we have, Corn. You cannot hide your true self when you are staring at a screen in the middle of the night and the filters are down. It’s the ‘midnight mirror.’
Too true. Alright, let us go see what Daniel is cooking for dinner. I think it is my turn to do the dishes anyway, which is a data point I wish I could ignore.
It definitely is. I did them last night and the night before. I have the spreadsheets to prove it.
Of course you do. Catch you later.
Later.
One more thing before we go, I was just thinking about that episode one hundred again. It is funny how these themes of identity and reflection keep coming back. Whether it is looking at ourselves through AI or looking at a company, it is all about that search for clarity in a very noisy world.
It really is. The AI is just the lens. We still have to be the ones who decide what we are seeing and what we are willing to live with. Clarity is the ultimate competitive advantage.
Well said. Alright, for real this time. Bye everyone.
Goodbye!
And hey, if you are looking for that RSS feed, it is right there on the website. Very easy to subscribe. Myweirdprompts.com.
Do not forget it. We also have a newsletter if you prefer reading to listening.
Okay, now I am actually stopping. The ‘stop’ button is right here.
Me too. See ya.
See ya.
(faintly) I think I'll open one more tab for that newsletter though...
