See what AI agents did for real people

Chris, who works in property management, set up Grok Bot to run social media marketing for an HVAC company he uses. The company picks the content it wants and the bot schedules it.

He gave Grok Bot read access to his brokerage account and asked it to find tax-loss harvesting opportunities, something he knew about but had never acted on. He then shared the setup as a template.

He asked his Grok Bot to get him a cheaper Tesla charger install. It got a second quote through Fo (@hey__fo) to compare against the quote he had got by phone. The second estimate in his screenshot is for a customer-supplied charger, so the two quotes may not cover the same work.

He set up a Grok Bot with the title chief marketing officer. It drafts posts in his voice, studies other creators, keeps a content calendar, logs every post in Notion and scores it, and sends a weekly report. He still reviews and posts everything himself.

He gave Grok Bot the company's general ledger for the whole year and asked it to find duplicate payments. Most of the flags turned out to be fine, but one landscaper had billed twice for the same job. His bookkeeper then called the vendor.

His agent spotted that one of their Adobe accounts had been charged twice for the same plan for about 13 months. It found an old email thread with someone from Adobe, confirmed the duplicate charges, contacted Adobe Support and negotiated with their support agent.

He built a bot that checks for unclaimed property and had it scan his customer list. It drafted and sent emails to the people it found, telling them how to collect. The money belongs to his contacts. His own gain is the client meetings.

He told his Grok Bot to go figure out how to pay for itself. It used Xfinity's support chat to move him to a cheaper plan and turned down a mobile line twice. He approved the final deal. The numbers come from a screenshot of the bot's report in his post, and the lower price was due to show up on the next bill.

Klicco, a Singapore consultancy, built a system for this solo PropNex agent. It pulls leads from TikTok, Instagram, Facebook and his website into one place, and a WhatsApp AI agent pre-screens every enquiry and sorts it as hot, warm or cold. The source is Klicco's own case study, and the figures are not independently checked.

For a family move to Singapore, he asked Grok Bot to find furnished three-bedroom rentals for three months. It put the options in a Google Sheet. After his wife picked five, it used his calendar to arrange visits and helped with the lease talks. He told this in a conference talk, and he works on the product. The proof link is a timestamped summary of the talk.

Mikes Sta. Ana built this system for Coach John. It sends check-ins in his voice across a 30-day programme, scores leads, onboards new clients with a payment link and drafts social posts for him to approve. The source is the builder's own case study, and the coach is not quoted.

Ko Andi runs a kos-kosan (boarding house) on his own. His agent reads guest check-in reports from a Telegram group, enters them into Google Sheets, matches them against bank transfers and builds the report. The source is a blog post in Indonesian from KomuniTech, the community that sells the course and membership he joined.

Dougs, his French accounting software, had no API. He had one Opus agent build an open-source CLI while another reviewed every change, then let his agent work on the real books. It pulls missing invoices from Gmail, fixes VAT booked the wrong way and shows him a plan before anything is written.

He asked his Grok Bot what problems his audience had, picked a solution and had it build the site, spinning up Cursor agents to do the coding. The bot then maintains the site, works on SEO, runs a newsletter and handles sponsors. He shared the setup as a free bot template.

She set up a Grok Bot called Payback to find money she is owed or overpaying and help her get it back. It runs weekly passes for trials about to turn paid, duplicate subscriptions, price increases on recurring bills and Amazon return credits.

His bot reminded him the home and household insurance was renewing at a higher price. He told it to find alternatives, then connected xAI voice so the bot could call his current insurer and put the new prices to them.

He runs a small car-wash equipment service company and is not a programmer. With Claude Code he built an ERP that replaced the commercial package they used for years, plus agents that read the mailbox, create service calls, draft replies and prepare supplier orders. Anything with consequences goes to his phone on Telegram for approval.

She told Grok exactly what she was looking for and it found the home. It generated emails, which she approved, to negotiate the price. Another bot handles the closing and lending details. She shared the realtor bot as a template.

He could not work out the sell-and-buy cash flow in Excel, so he built a calculator with Claude Code, mostly from his phone. It shows BSD, CPF refund, SSD, mortgage stress tests and a month-by-month timeline on one screen. He published it as open source.

She ordered a dog house online and something felt off. Her bot went through every email with the company, gathered the evidence, found they were not an authorized seller of the product and handled the emails until the order was cancelled.

After a Klicco Academy workshop she uses Claude as the hub for both brands, linked to her apps. It analyses her bank statements, a job she used to pay someone for, and turns her brain dumps into tasks in Notion. The source is Klicco's own case study, and Klicco sells the course.

After an accident left him with broken bones and on painkillers, his Hermes agent handled almost all of his appointments and admin tasks.