Stop chasing admin. Let AI agents do the work.
Custom AI agents that watch your systems, take the action, and only ping you when a human is needed. Built by an engineer with 20 years in data and automation.
- 20 years in data & automation
- First automation free
- Works in WhatsApp or Telegram
- Sydney-based
Add Priya to Saturday's 9amβ2pm shift, front desk
4:02 PM
Got it β add Priya Sharma to Sat 9:00amβ2:00pm, Front Desk. She's free that window. Confirm?
4:02 PM
yes
4:03 PM
β Done. Priya added to Saturday's roster and notified by text. Weekly roster updated.
4:03 PM
For Small Businesses
The same kind of automation, built for your operations β workflow automation, document extraction, dashboards, scheduled reporting.
See services →projectsMy Automations
Real projects running for my own household β portal scrapers, WhatsApp alerts, AI receipt extraction, a finance dashboard with its own chat assistant.
See the projects →marketsOne Dashboard, Every Broker & Wallet
A concept to bring every stock and crypto holding β brokerage account, exchange, wallets β into a single view, so checking your portfolio means one place to look, not five logins. Built alongside a research pipeline that already merges fundamentals, options data, and analyst ratings into one ranked report.
See the concept →Free new website. You bring the domain, I build it and host it free for 2 years.
Already have a site? Free redesign on the same terms. No obligation.
Own a small business? Your first automation is free.
Tell me what eats your week, and I'll build the first automation at no cost.
Every system feeds into one agent that acts.
Sales website leads, Facebook and Instagram marketing, scheduling, and bookkeeping β normally four different logins and four different apps. One AI agent watches all of it, checks the real system behind whichever question or instruction you give it, takes the action you approve, and reports back. We can build you a custom app for it, or just run it through WhatsApp or Telegram β apps your team already has open β either way, it does everything the same.
how many leads came in from the website today?
8:41 AM
π 14 new leads from the site today β 5 already got an automatic reply.
8:41 AM
how's the facebook campaign doing this week?
8:42 AM
π£ Facebook: 1,240 reach, 38 leads, $412 spent β best performer was the "Spring Special" post.
8:42 AM
book Meera in for 2pm tomorrow
8:44 AM
β Booked β Meera, Thu 2:00pm. Calendar updated and confirmation texted to her.
8:44 AM
how's spend looking this month?
8:45 AM
π° $8,420 spent against $34,150 revenue so far β synced to your accounting tool.
8:45 AM
What is manual work costing you?
Move the sliders to match your business and see roughly how much time automation could hand back each year.
A rough estimate to size the opportunity, not a quote.
Log an expense by texting it.
No spreadsheet, no shoebox of receipts β just say what was paid for. The agent logs it, proposes how to categorize it, and confirms before anything is finalized. Once confirmed, it's synced straight into your accounting tool β the work actually gets done, not just talked about.
See more business use cases →paid $340 for gym towels
11:14 AM
Logged $340 as a Supplies expense, paid from the business card. Want it tagged to a cost center?
11:14 AM
yes, tag it to Facilities
11:15 AM
β Done. $340 logged under Facilities and synced to your accounting tool.
11:15 AM
Hear it from Adam
Adam runs Repetitions Fitness. Here's what changed once the gym's admin was automated, in his words.
“15 hours a week of regular admin tasksβ¦ now they're all done.”
Adam, Owner ·Repetitions Fitness β
“Everything can be done now from the click of a button on my phone. I don't need my laptop to log in.”
“Something used to take me between an hour, an hour and a halfβ¦ now it takes like 30 seconds.”
“You're going to be able to save time, save money and ultimately save stress.”
Live for real clients
Not just concepts β here's what's actually running right now, and how far each build has gotten.
Personal Finance Dashboard
A private web app that pulls together every account, transaction, and spending category into one view β bank exports import automatically, spending rolls up on its own, and an AI chat assistant answers questions grounded in the actual data instead of generic advice.
See it in the portfolio →How much did I spend on groceries this month?
$612 across 14 grocery trips β up 8% from last month, mostly two big Costco runs.
One Dashboard, Every Broker & Wallet
Right now that's $100k sitting in IBKR, $30k in Binance, $20k in eToro, and $10k in Stake β four logins just to know where you stand. This pulls every balance into one total, updated in one place.
See the concept →Brokerage, exchange, and wallet balances β pulled into one total instead of four separate logins.
Where it started: my own household.
Before any of this was for clients, it was for my own family β a swim portal with no notifications, a solar inverter no one checked, receipts piling up unlogged. Small agents watch each one and report back the moment something matters, all through a single WhatsApp thread.
See the household projects →π Swim schedule this week: Anika β Tue 4:00pm (Level 3) Β· Arjun β Thu 4:30pm (Level 2). 1 makeup token for Arjun, expires Sunday.
7:02 AM
π§Ύ Woolworths receipt logged β $84.32 across 12 items. Added to groceries.
12:41 PM
βοΈ Solar: 6.2kWh surplus right now β enough headroom to start EV charging.
1:15 PM
Hi, I'm Kireeti.
I've spent 20 years in data and automation, building enterprise solutions: alerting, automated workflows, and the data pipelines behind them. I've also built quant options frameworks for SPX, with algorithms that buy and sell automatically.
The automations on this page started as fixes for my own household's repetitive chores, like checking portals with no notifications or typing up receipts by hand. Now I bring that same enterprise-grade approach to small businesses in Sydney and beyond.
More about the consulting side →



