Coverage
Every decision leaves a trace
Move your modelsonto the operating floor
Cirreo is an applied AI studio. We take systems out of the pilot stage and put them where the work actually happens — with throughput, cost and accuracy targets agreed in writing.
Queue cleared
$46,900
Intake92%
Review71%
Payout64%
Eval run 148
Accuracy held through peak week
Method
Strategy, data and shipped software from one team
Model care
Retraining on your own labelled examples
Docs parsed
520k+
Q1↑ 18%
Q2↑ 24%
Handled end to end
69%
Northmoor Trust
Calder Health
Halyard Freight
Ironvale Works
Basten Retail
Meridian Group
About us
A consulting partner built for the messy middle — where a promising model meets a real process, a real ledger and a real team
Cirreo
120+
Systems in production across cloud, on-premise and air-gapped environments.
Targets met on contract 100%
They rewrote how our claims desk works. Quiet, fast, and nothing broke on the way in.
Contracts, claims, invoices and field reports — read, routed and checked against your own rules.
What we do
Six things we are asked for, over and over
Each one is scoped as a standalone piece of work. Most clients start with one and add the next once the numbers hold.
Opportunity mapping
We sit with your team for two weeks, watch the work, and come back with a ranked list of what is worth automating and what is not.
Model selection and tuning
Open weights, hosted APIs or a small task-specific model. We test candidates against your data before anyone signs a licence.
Workflow engineering
The unglamorous part: queues, retries, human review steps and the handoffs that decide whether anyone actually uses the thing.
Governance and controls
Audit trails, access rules, red-team runs and a written record of what the system may and may not decide on its own.
Evaluation and monitoring
A test suite built from your real edge cases, running on every change, with alerts when quality drifts instead of after a complaint.
Handover and training
Your engineers run it after we leave. Documentation, pairing sessions and a 90-day window where we stay reachable.
Delivery path
Four stages, in this order, every time
Order matters here — each stage produces the input the next one needs. We do not skip ahead, and we stop at any gate that fails.
Ground truth
We measure the current process by hand: volume, cost per item, error rate, where time goes. Nothing gets built against a guess.
Weeks 1–2Narrow prototype
One workflow, one team, real data. The goal is a decision, not a demo: is this worth the integration cost or not.
Weeks 3–6Production build
Hardening, permissions, monitoring and the review steps your risk team asked for. Rolled out to one site before all of them.
Weeks 7–14Handover
Your team takes the keys. We shadow for a month, answer questions for two more, then step back for good.
Weeks 15–18
4.2×
median return on the first workflow shipped
How we work
We stay until the numbers move, then we leave
Most AI work stalls somewhere between a convincing prototype and a Tuesday morning shift. Our whole practice is built around that gap.
-
One team, start to finish
The people who write the strategy write the code. No handover to a delivery pod you have never met.
-
Targets written into the contract
Cost per item, cycle time, accuracy floor. If we miss them, the last invoice adjusts.
-
Your data stays yours
Deployment inside your account or on your hardware. Nothing trains on your records without a signed instruction.
-
No lock-in by design
Standard tooling, documented interfaces, and a swap path for every model we recommend.
Sectors
Where our reference cases live
We take work outside these six, but slower — we will tell you if the learning curve is on your budget.
Banking and insurance
Claims triage, document review, adverse media checks and the audit trail your regulator will ask for.
31 engagementsHealth systems
Prior authorisation, coding support and clinician-facing summaries with a named reviewer on every output.
18 engagementsLogistics
Exception handling across freight documents, customs paperwork and carrier correspondence.
22 engagementsIndustrial
Maintenance history search, defect classification and shift-handover notes that survive the handover.
14 engagementsRetail and marketplaces
Catalogue enrichment, fraud review queues and support desks that resolve without a scripted loop.
19 engagementsPublic sector
Casework backlogs and correspondence handling, built for records requests from day one.
9 engagements
Measured, not estimated
What the last three years produced
Figures come from client-side reporting at the six-month mark, not from our own dashboards.
90
Engagements delivered since 2019, across 20 countries.
69
%
Median share of a queue handled without a person touching it.
11
wks
Average time from kickoff to the first workflow running live.
4.2 ×
Median first-year return on the initial build cost.
Selected work
Request the full dossier
Three projects, described plainly
Insurance
Clearing an 11-week claims backlog
Northmoor Trust had 40,000 motor claims waiting on document review. We built an intake reader with a two-tier human check and rebuilt the routing rules underneath it.
11 → 2weeks to decision
64%auto-cleared
Logistics
Customs exceptions, handled overnight
Halyard Freight lost margin to paperwork mismatches caught at the border. The system now reconciles documents before dispatch and flags what a broker needs to see.
$2.4mheld detention cost
92%caught pre-dispatch
Industrial
Forty years of maintenance logs, searchable
Ironvale Works had repair history spread across scanned binders and three retired systems. We indexed it and put answers in front of technicians on the floor.
38%faster diagnosis
1.2mrecords indexed
Client words
What people say once we have gone
All of these were collected at the six-month review, not at go-live.
They spent the first fortnight watching our adjusters work and asking questions. Everything after that was built for the desk we actually have, not the one on the org chart.
Priya Raghunathan
Head of Claims, Northmoor Trust
We had been burned by a pilot that never shipped. Cirreo scoped the smallest useful thing, put it live in nine weeks, and let the results argue for the rest.
Tomas Berglund
COO, Halyard Freight
The handover was the part I did not expect. My engineers can change the system without calling anyone, which is the only reason it is still improving.
Adaeze Okonkwo
CTO, Ironvale Works
The eval suite is the thing that changed the argument internally. We stopped debating whether it worked and started reading a number every morning.
Marcus Feld
Director of Ops, Calder Health
They told us one of the three workflows we asked for was not worth building. That is the moment I decided to keep working with them.
Yuki Tanaka
Head of Data, Basten Retail
Everything sits in our own cloud account with our own keys. Our security review took two weeks instead of the usual six months.
Ana Ferreira
CISO, Meridian Group
Engagement models
Three ways to start
Fixed scope, fixed fee. Anything outside the scope gets quoted before it starts, never after.
$18k
fixed, three weeks
For teams who need to know whether there is a case at all, and want a number they can take to a board.
- On-site process study
- Ranked opportunity list with costs
- Build-or-buy recommendation
- Board-ready written report
$95k+
fixed, 12 to 18 weeks
One workflow taken from ground truth to production, with targets written into the statement of work.
- Everything in Assessment
- Prototype, then production build
- Eval suite and monitoring
- Documentation and pairing
- 90 days of post-handover support
$34k
per month, six-month minimum
For organisations running several workflows at once who want a dedicated group instead of a queue of requests.
- Named team of three to five
- Quarterly roadmap you set
- Model and vendor review
- Training for your engineers
Insights
See all writing
Notes from the delivery floor
Your pilot succeeded and nothing changed. Here is why.
The gap is rarely the model. It is usually the eight steps around it that nobody costed. A pilot proves…
admin · 1 min read
A smaller model, tuned on your own examples, usually wins
What we found running the same six tasks across hosted and open weights for a year. On narrow, repetitive tasks…
admin · 1 min read
Write down what the system may decide alone
A one-page decision boundary saves more arguments than any policy document we have seen. Before the first line of code,…
admin · 1 min read
Questions
Asked before every kickoff
If yours is not here, write to us and we will answer it in plain language.
No. Waiting for clean data is the most common reason projects never begin. We work with what exists, document what is missing, and fix the records that actually affect the outcome.
Inside your cloud account, your data centre, or an isolated environment with no outbound network. We have delivered all three. The choice is yours and it does not change our fee.
The final invoice reduces on a scale written into the statement of work. We would rather agree that up front than argue about it at the end.
Only on your written instruction, only on your infrastructure, and only for a model you own. Vendor terms are reviewed and summarised for you before anything is signed.
We prefer it. Projects with two or three of your engineers embedded hand over faster and hold their results longer. It also reduces the fee.
Assessments usually begin within three weeks. Build engagements depend on team availability — as of this quarter, the next opening is in about six weeks.
Next step
Bring us the process nobody wants to touch
Send a short note about the workflow and the number you need to move. You will get a written reply from an engineer, not a form response, within two working days.