What we build
Three lines of work. They overlap more often than not, and most operations need a bit of each. Every one of them ends with something you own.
AUTOMATION
Your tools stop needing a human courier.
We wire the systems you already pay for so they hand work to each other directly. Triggers, routing, validation, retries, and the unglamorous error handling that decides whether an automation survives contact with a real month.
- iCRM and database entry, written at the moment of the event
- iiInvoices, reports and documents that generate and file themselves
- iiiLead routing and follow-up that never waits in someone's inbox
& RETRIEVAL
An answer desk that has read everything you own.
Agents grounded in your actual contracts, tickets and policies rather than the open internet. They answer, qualify and triage, and hand over to a person the moment the answer isn't in the record instead of inventing one.
- iCustomer and internal support that cites the document it used
- iiLead qualification and enrichment before the first human call
- iiiSearch across every contract, ticket and wiki page you hold
INTERNAL SOFTWARE
The tool your team keeps asking for, built to your shape.
When nothing off the shelf fits, we build the thing itself. Dashboards, portals, approval flows, the console the operation actually needs. Yours to keep, documented, no per-seat licence, no vendor holding the keys.
- iInternal dashboards and operational consoles
- iiClient portals and multi-step approval workflows
- iiiIntegrations between systems never meant to speak to each other
How it goes
Four steps from the first call to a system that runs without you. We move in days rather than quarters, and you see the scope and the price in writing before we write a line of code.
You show us what eats the week
The Monday morning dread, the spreadsheet everyone is quietly afraid to touch. We ask an uncomfortable number of questions and take a great many notes.
Half an hour, one person who actually does the work. No prep, no deck, no tidying up first. The mess is the useful part.
We come back with the map
What can be automated, what shouldn't be, what it costs and what it returns. A fixed price and a fixed scope, in writing, before anything is built.
An hour to read it and argue with it. If the answer is that this isn't worth building, that is a real answer and you get it in writing too.
We build it against your real data
Your real edge cases, not a demo dataset. The awkward client, the malformed export, the month everything arrives at once. You watch it take shape and tell us where it is wrong.
Access to the systems, and roughly two hours a week to look at what we've built and say what's wrong with it. That's the whole ask.
It goes into production and your week opens up
We watch it, repair what drifts, and adjust as the operation changes, because it will. Handover documentation either way, so you are never holding a system nobody understands.
Tell us when the business changes shape. Everything else is ours until you'd rather it wasn't.
You get one number for one scope. If it takes us longer than we thought, that is our problem and our cost, not a bigger invoice.
The price arrives after the mapping, not before it. Anyone quoting you on the first call is guessing, and you pay for the guess later.
No per-seat licence, no platform fee, no vendor holding the keys. It runs on your accounts and it is yours the day it ships.
The care plan is monthly and you can stop it. You keep the system, the documentation and the credentials either way.
The awkward questions
The ones that actually decide it, answered the way we'd answer them on the call.
Why not just use ChatGPT?
For drafting an email, use it. The gap is that it doesn't run when you're asleep, it can't reach into your CRM, and it has never read your price list. What we build is the plumbing around a model, and on plenty of jobs there is no model involved at all.
Why not hire someone cheaper to do it?
A person doing data entry costs the same next year and the year after, needs managing, and leaves. Sometimes hiring is still the right answer, usually when the work needs judgement rather than repetition. We'll say so if that's what we find.
Who owns what you build?
You do. It runs on your accounts, under your credentials, with the source and the documentation handed over. We don't host it hostage and there is no platform you have to keep paying us for.
What happens when it breaks?
It will, because the systems it touches change under it. That's what the care plan is for: we watch it, we get alerted before you notice, and we repair drift. Anything that breaks because we built it wrong is fixed at our cost, plan or no plan.
What if you disappear?
Fair question to ask a small team. Nothing runs on infrastructure we control, so the worst case is that you hire someone else to maintain ordinary code on your own accounts. The handover documentation exists from day one, not from the day you ask for it.
What if it turns out it can't be automated?
Then the mapping says so and you've spent a few days finding out instead of a few months. We would rather hand back a document explaining why not than build something that quietly makes the work harder.
Show us the thing you dread on Monday morning.
Thirty minutes, no deck, no pitch. You describe the work that eats the week and we tell you whether there's a system in it.
We take on a small number of operations at a time, so the people who build your system are the people you spoke to on the first call.
If we can't help you, we'll say so on the call and point you at whoever can.
Use the form below, or write to us directly.
No newsletter, no drip sequence, no follow-up from a tool.
One reply from a person, usually within a working day.
We have it. Someone will write back within a working day, usually with a couple of times to choose from.
If it's urgent, or you'd rather not wait, email desaigpratik@gmail.com directly.