Time is Irretrievable — and Vibe Management is the Only Way to Stop Wasting It on What Machines Do Better

Monday, 8:00 AM. Before you even have a chance to open your email, your AI assistant has already scanned the latest updates across your company. It has reviewed metrics on your behalf, identified a bottleneck in the lead generation department and prepared three courses of action. You choose the second option and go grab a cup of coffee. The day has begun.

This is exactly what I call Vibe Management. And it is precisely where we are headed, faster than most executives can even realize.

From one term to another

You have likely heard of Vibe Coding — the practice where individuals without a technical background write functional code simply by conversing with an AI. I do this myself, and it works. The exact same logic applies to management: everything that used to require hours of administrative labor — assigning tasks, reconciling metrics, tracking financial threats and gathering data before strategic decisions — can now be handled by artificial intelligence. However, this immediately brings us to a crossroads that few people ever mention.

Weakness of a smart assistant

Try asking any neural network right now: “Who should I hire first?” or “Where is the bottleneck in my company?”. It will give you a generic, useless response. Not because it is stupid, but because it is “starved.” The AI does not know who is responsible for your conversions, who is handling key negotiations or which metrics you consider critical in the first place.

Without live data regarding your organizational structure, functions, and the specific metrics of actual people, even the most powerful model remains nothing more than a glorified calculator. And if you attempt to upload this data manually every single time, the endeavor turns into torture, consuming more time than it saves.

This is exactly why we have spent several years building our own management platform. The organizational structure there is not just a pretty diagram drawn on a wall — it breathes. Inside each node live actual functions, linked metrics and a history of results. When an AI connects to a database like that, it stops being just a chatbot and becomes that very analyst you have long wanted to hire but could never afford.

What this means in practice

I am not talking about theories. Here are two tools that are already reshaping my work every single day.

• Financial Assistant: I have a Claude-based assistant with access to accounts, cards, and liabilities. Before any major purchase, I ask a single question and receive a well-balanced response backed by figures, pointing out illogical transactions and warning of potential cash flow gaps. It doesn’t just give a “yes” or “no” answer; it explains why.
• Investment Proposal Analyzer: The AI assistant understands my values, interests, and approaches to project evaluation. When someone sends over a pitch, I simply forward it to the assistant. It first looks for internal contradictions in the data, and then, if necessary, handles the correspondence with the authors themselves to request missing information. If the numbers add up and the logic holds up against market scrutiny, it prepares the agenda and structure for our discussion.

This saves an immense amount of time. Massive investment spreadsheets that used to require hours of deep analysis are processed by the assistant in minutes. Crucially, even if a project is not a good fit, every author receives substantive feedback from my assistant: what to focus on and what to refine so the idea can truly take off. I am glad when people send me their projects, and I want to provide genuine value in return.

This is neither science fiction nor startup hype. JPMorgan Chase uses its COiN platform to finalize the analysis of commercial loan agreements in seconds — a task that previously consumed 360,000 hours of lawyers’ time annually.

How to prompt your team to embrace new order of things

There is no magic to it. If you expect employees to adopt AI tools on their own, it might never happen. My partner and I went through this and developed an approach that actually works.

1. Lead by Example: First and foremost, the founder must be a living example, rather than just decreeing that “we are implementing AI.” People look at an owner not as an instruction manual, but as a real person: if they use it themselves, talk about it, and showcase results, it becomes more contagious than any corporate mandate.
2. Remove the Technical Barrier: Purchase access to paid models and make the process seamless and user-friendly.
3. Change Deliverable Standards: For tasks involving analysis or document preparation, we began requiring not just the final output, but also the prompt used to generate it. It sounds like a minor detail, but when you read someone else’s prompt, you see how they think and understand why the report turned out the way it did.

In parallel, we launched dedicated internal chat rooms, not about general AI market news, but focused on the specific tools we use: HeyGen for video, ElevenLabs for audio and so on. A designated team member handles “lab work,” meaning they test new tools before they are introduced into the workflow.

We also run hackathons. We gather key people in one space for a set period, present concrete business challenges and solve them in real time, discussing and finding solutions together. Our first hackathon yielded a breakthrough for our accelerator program: right then and there, we mapped out the functional structure and data collection logic. Today, hackathons run continuously within the company; participation is open to any employee and real rewards are given for winning and implementing new tools.

How the leader’s role will change

The question everyone asks is: if AI controls everything, why bother understanding management at all? The answer is the same as in vibe coding: if you do not understand the building blocks your system is made of, you will quickly hit a ceiling.

However, the nature of the work is transforming. Administrative control in the form of reconciliations, checks and reminders is shifting to algorithms. Hierarchies will not disappear, but they will become flatter: a single leader will be able to oversee teams that previously required multiple layers of management. A leader’s job will not get easier; it will become different, more strategic and more human, in the best sense of the word.

Where to start today

Three steps, without which everything else is meaningless.

• First, the Data Foundation: Create a living database: who is in the company, what they are responsible for and which metrics are tied to each role. Without this, AI will be giving advice blindly.
• Next, Task Control: Start with the simple things: voice commands, automated reminders and deadline tracking. Feel your mind clear up.
• In Parallel, Information Filters: I am currently building a personal “news curator” — an assistant that filters out informational noise and leaves only what truly impacts my business decisions. Time is irretrievable and Vibe Management is, above all, a way to stop wasting it on things a machine can do better than you.