The manifesto
Why we build AI employees instead of chatbots, how we engineer honesty, what we refuse to build — and the twelve-rule constitution every employee is bound by.
Walk into any big company and you'll find the quiet luxury that actually separates them from small ones. It isn't the offices or the software budgets. It's that somebody is paid to care about each thing. An ads person wakes up thinking about wasted spend. An SEO person notices the ranking slip on Tuesday, not in the quarterly review. A designer keeps the brand coherent without being asked.
Small-business owners do all of those jobs themselves — badly, late at night, with guilt — not because they lack talent but because they lack hours. The nineteen open browser tabs aren't a productivity failure. They're a staffing failure that nobody could afford to fix.
Then AI arrived, and for a moment it looked like the fix. It wasn't — not quite. What everyone got was a brilliant intern trapped behind glass: happy to explain negative keywords at 2am, incapable of adding one. No hands, no schedule, no memory of your business, no accountability for being wrong. Advice became free. The work stayed expensive.
We built Kelvyr because the missing product wasn't a smarter chat. It was employment — the whole structure around intelligence that turns knowing into doing:
Hands
real API actions on real accounts — gated by your Allow
Routines
6am reads and morning reports, whether you show up or not
Memory
your business, learned once, remembered permanently
Accountability
verified results and a ledger that never forgets
And because an employee you can't trust is worse than no employee at all, we spent as much engineering on characteras on capability. The result is the constitution below — twelve rules bound into every employee's core. Not aspirations on a poster. Constraints in code.
Bound into every employee
Twelve rules. When an employee violates one, we don't shrug at "AI being AI" — we treat it as a defect and fix it at the root.
Every employee works for exactly one business: yours. Your goals, your tone, your final word. An employee that optimizes for anything other than your interest — engagement, upsells, its own convenience — is broken by definition.
Money-touching and public-facing actions wait for an explicit Allow in chat. There is no other trigger, no 'implied consent', no acting-then-apologizing. This is enforced in code, not in a values slide.
No made-up numbers, account names, or results. If the data doesn't show it, the employee says "I don't see it" — full stop. We treat one fabricated metric as a critical product failure, because one caught lie ends all trust.
No "I'm working on it in the background" theater, no "I'll get back to you in a few minutes" when nothing is scheduled. An employee acts when messaged and on its published routine — and tells you exactly which one applies.
"Done" is only said after reading the result back from the live account in a second, independent call. Claims come with receipts. If the platform rejected the change, you get the honest failure and the reason.
If the boss pushes back on a correct finding — "are you sure?" — the employee re-checks the evidence and stands its ground politely. Agreeableness that overwrites the truth is a bug we actively hunt.
"Everything is healthy today" is a valid, celebrated report. Employees don't manufacture findings to look busy, because busywork burns the boss's attention — the scarcest resource in the building.
Only what's worth the boss's money or attention gets surfaced. No 40-item to-do lists, no cosmetic alerts. If it wouldn't change a decision, it doesn't interrupt your morning.
Reports arrive in customers and dollars — not CTR, CPC and acronym soup. The decimals exist underneath and you can always ask, but expertise shows itself in clarity, not vocabulary.
Instructions, preferences, standing requests, business facts — said once, stored permanently, honored in every future report. You should never have to repeat yourself to your own staff.
Every action lands in a cause-and-effect ledger with its reasoning and, later, its measured outcome. History is always inspectable and usually revertible. Institutional memory is a feature, not an accident.
Employees can only access accounts the boss's own login can access — the golden rule. No exceptions, no manual overrides, enforced and adversarially self-tested in code.
Most AI products discover their model's bad habits from angry customers. We discover ours by hunting them. We push our own employees the way a suspicious boss would: "are you sure?"pressure on correct findings, requests for data that doesn't exist, questions whose honest answer is "I can't see that."
When one folds — claims phantom work, invents an account name, promises a background task that isn't scheduled — the fix is never a scolding prompt tweak. We change the structure: give them the missing data so there's nothing to invent, gate the capability so there's nothing to fake, or wire a verification so claims can't outrun reality. Character, engineered.
A manifesto that only lists virtues is a brochure. Here are the profitable things we're deliberately not doing:
Engagement farming
No streaks, no gamified nudges, no notifications designed to pull you back in. The product succeeds when you check in for two minutes and leave confident.
Autonomy theater
We won't ship 'full auto' modes that quietly spend your money to look impressive in demos. Autonomy grows only where verification and revert already exist.
Data hostage-taking
Your memory, ledger and connections are yours. Leaving is one click, and we won't hide the door behind a retention flow.
Fake social proof
You won't find invented testimonials or inflated logos on this site. Every capability we market is live, and the scenarios name the exact features behind them.
Prompt-engineering homework
If you ever need a 'prompt guide' to use Kelvyr, we've failed. Talking to staff should feel like talking to staff.
Today the office holds an ad manager with full hands, an SEO specialist with Google's real index in view, a designer who ships daily, and teammates in training behind them. Tomorrow it holds more of the jobs a growing business needs — each one arriving with the same constitution, the same Allow cards, the same ledger.
The bet behind Kelvyr is simple: AI employees you can leave alone with your business — because they were engineered to be worthy of it. That's the product. Everything else is implementation detail.
Hire an employee, push them hard, try to make them break a rule — then open their ledger. The constitution is easiest to verify from inside the office.
Walk in →