Solve · HR
Most of the HR inbox is the same twenty questions. Holiday, sick pay, probation, expenses, the parental policy, who signs what. The answers already exist, written down, in a handbook nobody opens. J Labs builds an assistant on your own policies that answers them, shows the source for every answer, and hands over to a person when the answer is a judgement.
Summary
J Labs builds assistants on a company's own HR policies, for people teams whose inbox is mostly the same recurring questions. One defined body of material is used, the handbook, the policy set and the standard letters, and nothing outside it. Every answer carries the section it came from, so the person asking can check it and the HR team can see what the assistant is telling people. Where the material does not cover the question, the assistant says so and hands over, rather than producing a plausible answer that somebody then acts on. The boundary is written into the scope before anything is built: questions about one named person, about pay, performance, grievances or dismissal go to a human every time, because those are decisions a company owns and not content to be retrieved. When a policy changes, the assistant changes with it, because it reads the policy rather than a copy of it. It runs inside the chat tool your team already uses. Engagements are fixed scope, fixed timeline and fixed price, from the Spec at €5,000.
What we automate
One defined corpus, agreed up front: the handbook, the policy set, the standard letters. The assistant does not reach for general knowledge about employment, because general knowledge is not your policy.
Each answer quotes the section it came from. The person asking can check it, and the people team can see what is being said without reading every conversation.
Where the policy is silent, the assistant says so and hands over. This is the single most important behaviour and the one most products get wrong, because a confident wrong answer about sick pay is worse than no answer.
Anything about one named person, pay, performance, a grievance or a dismissal is routed to a person rather than answered. The list is agreed in the scope, not decided by the model.
The assistant reads the policy rather than a copy of it, so updating the document updates the answers. No retraining, and no second version of the truth to keep in step.
Slack, Teams or the intranet. A question answered where people already ask it gets asked; a separate portal with a separate login does not.
Fit
This is for you if
Your policies are written down and reasonably current. They do not have to be perfect. They do have to exist, because an assistant on your policies is only as good as the policies, and this work will show you exactly which ones contradict each other.
A real part of the HR inbox is repeat questions. The same twenty questions, asked by different people, answered by whoever picks up the inbox. That is the shape where this pays back.
You can name the questions it must handle. A known set of questions against a defined body of material. That is the shape the Spec covers: specified, built and tested for a fixed price.
This is not for you if
You want it to decide individual cases. Hiring, performance, pay and dismissal stay with a human on every engagement. An assistant that rules on a person is a liability, not a saving, and we will not build it.
You want it to give employment law advice. It answers from your policy. Where a question turns on the law rather than on what your handbook says, the honest answer is a person and, often, a lawyer.
The policies do not exist yet. If the answers live in three people's heads, write them down first. That is a genuinely useful piece of work and it is not this one.
Systems
The ones you already run. We don't insist on rip-and-replace: the automation layer goes on top of the systems in place, and if a system is genuinely past saving we say so on the discovery call and show our working.
Engagement options
| Engagement | Price | Timeline | What you get |
|---|---|---|---|
| The Spec | €5,000, fixed | Fifteen working days | One clearly defined outcome, designed, built and tested: one body of policy material, a known set of questions, sourced answers, the hand-over rules and the human boundary. A free assessment confirms it fits before you pay. |
| HR operations build | Fixed price, quoted after discovery | Typically four to eight weeks | The assistant plus the administration around it: onboarding and offboarding sequences, document generation from the HR record, and the approval routing that currently runs on email. |
| Bespoke | Multi-phase, to €200,000+ | Multi-month, phased | People operations across entities or countries, where policy, language and legal context differ by location, phased with weekly check-ins. |
None of these is an off-the-shelf product. There is no J Labs software to install: every system is specified in the brief, built for your stack, and owned by you. Every engagement is fixed scope, fixed timeline and fixed price, with acceptance criteria agreed before work starts. See how we work and the Spec.
The product that fixes it
Private Team AI Workspace · 2,450 EUR setup + 395 EUR / month
A workspace on your own documents, with access boundaries between teams, is the productised version of this: your material, your people, sourced answers, and no data going anywhere you did not agree.
Live availability
Live slots from our team calendar, weekdays, European working hours. A 20-minute intro or a 45-minute working session, your choice, on Google Meet. You get a one-page summary within 24 hours.
How much time do you need?
Step 1 · Pick a date
Loading availability…
Frequently asked
Or start with our brief templateWe reply within one business day.