How one founder turned “Jerry” into a full-time AI personal assistant

Imagine having a tireless, ultra-organized employee who never sleeps, manages your inbox, makes calls for you, summarizes your day, and even analyses your YouTube channel — all powered by AI. That’s exactly what one entrepreneur has built with an AI assistant he calls “Jerry”.

Treating AI Like a Smart Employee

The core mindset behind Jerry is simple: treat the AI like a very smart employee, not just a chatbot. Instead of constantly asking “Can it do this?”, the owner assumes Jerry can handle almost any digital task and then builds workflows around that assumption.

Jerry is wired into tools, APIs, and browsers so it can access data, take actions, and report back in a structured way. Over time, it has evolved from a basic helper into a full-blown personal operations system.

How Jerry Manages Email So Nothing Gets Missed

One of the biggest problems this founder faced was missing important emails because the team primarily works on Slack. To fix this, Jerry now acts as an intelligent email gatekeeper.

Here’s how the workflow is set up:

• Inbox scanning every 2 hours: Jerry checks the email inbox regularly using tools like Google CLI to access data from the backend.
• Detecting important emails: If Jerry thinks an email is important, it sends a notification on Slack.
• Escalation if there’s no response: If there’s no reply on Slack within an hour, Jerry pings on WhatsApp. If there’s still no response, it escalates further by placing an actual phone call.

If there are no urgent emails, Jerry simply sends a summary every six hours: a five-line digest with links to all recent emails. This keeps the inbox under control without constant context-switching.

Calendar, Travel, and Voice-Based Instructions

Jerry doesn’t just manage email — it also handles calendar and scheduling end-to-end. When the founder is traveling, for example to Hyderabad for two days, he can simply send a voice note describing the dates, timings, and rough plan. Jerry then turns that into a structured schedule and manages the calendar accordingly.

The key is natural interaction: instead of filling forms or clicking through interfaces, the owner just talks to Jerry in plain language, and the AI handles the rest.

AI That Makes Phone Calls For You

One of the most impressive parts of Jerry is its ability to make calls on behalf of the owner. This is powered by a combination of:

• A calling tool called Wappy, which provides an API for programmatic phone calls.
• ElevenLabs (11 Labs) for voice cloning, so the AI speaks in the owner’s voice.

With this setup, the owner can simply say something like, “Call and tell him I’ll be 30 minutes late,” and Jerry will:

• Place the call using Wappy’s API.
• Speak in a cloned voice that sounds like the owner.
• Follow a predefined script or “playbook” depending on the situation.

There are also detailed follow-up rules. For example, if the person doesn’t pick up:

• Call again after 5 minutes.
• If still no answer, call again after 30 minutes.
• If there’s an email or WhatsApp contact, send a message there instead.

Urgency tags guide the behavior. A highly urgent task (like informing someone about a delay) is handled differently from a casual check-in (like asking, “Hey, do you want to eat anything? Let me know.”). Jerry can even call back later with updates, such as confirming an order or sharing the other person’s response.

AI-Generated Content and a Consistent Voice

All of the owner’s social media content is AI-generated, and thanks to voice cloning, Jerry can speak in his voice across different channels. When Jerry talks, it sounds like the owner, but it clearly identifies itself as an AI assistant so that people who know him well can tell the difference — while others may not notice at all.

Automated YouTube and Data Analysis

Jerry is also used as a data analyst for YouTube performance. Instead of manually digging into analytics, the owner simply gives high-level instructions like:

• “Figure out what is working on YouTube for me.”
• “Find outlier content ideas I can create, using my VidIQ account.”

Because Jerry is connected to the necessary accounts and tools, it can:

• Open YouTube in a browser on the home computer.
• Download analytics reports if needed.
• Pull additional insights from tools like VidIQ.
• Run its own analysis and extract patterns.

Importantly, Jerry has been trained with a habit: whenever it presents data, it must build a dashboard, not just send a wall of text. That means the final output is visual, structured, and easier to act on.

If you’re interested in building something similar, you may find it useful to explore how to create your own agent system in this guide on building a personal agentic operating system.

What Does It Cost to Run an AI Assistant Like This?

So what’s the price of having an AI “employee” like Jerry?

For this particular setup, the current cost is around 1 lakh INR per month. That sounds high, but there are reasons:

• The owner is constantly experimenting and pushing the limits of what Jerry can do.
• Many different APIs and tools are being tested in parallel.
• This is more of an edge-case, power-user setup than a typical deployment.

For a more standard user, the estimated cost would be far lower — likely in the range of 5,000 to 10,000 INR per month, depending on usage. Most of the expense comes from API calls to large language models and third-party services.

Beyond API costs, you mainly need:

• A computer or laptop that stays on and connected.
• A framework or platform (like OpenClaw or similar) to orchestrate the workflows.
• Some initial setup time to connect tools, define playbooks, and set rules.

If you want a more structured walkthrough of building your own assistant, check out this step-by-step OpenClaw tutorial for creating an AI assistant.

The Future: AI as a Team of Digital Employees

Jerry is a glimpse of where personal AI is heading: not just chatbots, but full-fledged agents that behave like a team of smart employees. They monitor your inbox, manage your time, make calls, analyze data, and present insights in dashboards — all while you sleep or focus on higher-level work.

The real unlock is the mindset shift: once you start treating AI as a capable colleague and design workflows around that, you can offload a surprising amount of your daily digital work to an automated system.