A unit that already works together, pointed at your job.

Hand over a piece of work

A product, a platform migration, a new business line. The unit takes the whole thing, from the decisions down to the release, and you deal with one accountable person rather than a delivery organisation.

• One contract and one point of accountability
• A written picture of the state of things, kept current
• Handover to your people planned from the start, not at the end

Add weight to the team you have

Your team is good and simply outnumbered by the roadmap, or it is missing one capability that is holding everything else up. The unit works next to yours rather than instead of it: same repository, same standups, same definition of done, and no parallel project with its own version of the truth.

Sitting next to people who have done the thing before is also how developers get better at it, which is the part you keep. The unit should leave your team stronger than it found it.

• Your leads keep leading; nobody arrives to take over
• One backlog and one definition of done, not a project running beside yours
• Pairing and review both ways, so the capability stays after we go
• Your developers learn our agent workflow by using it with us, on real tickets
• Two people or five, and the number can change as the work does

Scoping before anything is promised

We go through the job properly: what has to be true when it is done, what already exists, what the constraints are, and which parts nobody can size yet. That last category is the one that sinks fixed bids, so it is named out loud rather than hidden in a number.

• What is in, what is out, and what is still unknown
• An honest read on whether this needs a unit at all

Paid the way the work actually behaves

The parts we can scope get a fixed price. The parts nobody can honestly size yet are billed by the hour until they are understood, and then they can move to a fixed price too. You are never paying a padded estimate for uncertainty, and I am never quietly absorbing it.

• One rate per capability, with everything included
• No vendor commissions, so a recommendation has only the work behind it

The people are named, not promised

You see who is coming, by name, with what they have delivered and when they are free. Availability is confirmed before there is an offer, so nobody is sold to you twice and no bench is invoiced to you.

• A named deputy lead on every engagement
• The mix can change mid-project without a new procurement

Working, and planning to leave

Three to twelve months is the usual shape, and a normal start is two to four weeks out. The handover plan starts on the first day rather than the last, because a unit that cannot be got rid of is a liability, not a service.

Weeks instead of months

Recruiting one experienced engineer is a three to six month exercise before anybody writes a line of code, and then they still have to learn the team. This is two to four weeks, and the unit can shrink again when the work is done.

No onboarding between the people

A team assembled from strangers spends its first months finding out who is good at what and how to disagree productively. This unit did that years ago, on other people's projects, so that time goes into your job instead. Your domain still has to be learned, and that part is honest work.

Agents are how the unit works, and your people are taught the same

Agents write, review and test code here every working day, so the pace already assumes them instead of treating them as a pilot. That is only worth something to you if it survives our leaving, so the rules and pipelines are written into your repository rather than kept in our heads, and your developers are shown the parts that are easy to get wrong.

• Agent rules and review practice written down, in your repository
• Automation from alert to ticket to pull request
• Quality and security gates that run without anybody remembering them
• Where not to point an agent, and how to tell when it is confidently wrong

One person to talk to, who is also doing the work

Problems get said once, to somebody who can act on them that day and who is in the code rather than above it. I take no vendor commissions, so a recommendation has nothing behind it but the work.

Your own team is better off afterwards

Nobody learns much from a supplier working behind a wall. Because the unit sits inside your repository and your reviews, your developers pick things up in the ordinary way people do, by working next to somebody who has done it before. When the engagement ends, that part does not leave with us.

Contracting and compliance

The contract is with my company, whatever the size of the unit. Its members are subcontractors on back-to-back terms, and they are paid when the client has paid.

• Tilaajavastuu and Valtti in order, and the report is here
• The same papers collected from every subcontractor
• Liability limits in the contract, and consultancy liability insurance

Pulling people off other work costs money

Nobody is sitting on a bench waiting, which is the point: these are people with work to do. They can be moved sooner, and sometimes much sooner, but breaking existing commitments carries a premium and I will name it plainly rather than bury it in a day rate.

A normal start is two to four weeks. Faster is a commercial question, not a capability one, so ask and you will get a straight answer and a number.

What it will not do is pretend

If your production is down this afternoon, the honest answer is that your own on-call rota will beat any outside unit to it, and I would rather say that now than after a contract.

If the fire keeps happening, that is a different job, and one this unit is well suited to: finding out why, and making it stop.