The standard
The Conntal Vetting Standard
Every Conntal candidate is screened by a practitioner at Ace8, AceMQ or Duczer East who runs that technology in production for enterprise clients. Recruiters source; engineers decide. The screen is a real production scenario, and we publish it.
The funnel
Five stages, and only one of them belongs to a recruiter.
We do not publish pass rates yet. We will when Conntal has processed enough candidates for the number to mean something — quoting a percentage before then would be exactly the kind of unverifiable claim this page exists to replace.
- 01
Intake
A practitioner reads the brief
Before any sourcing, an engineer who runs this technology reads your requirement and tells you what is wrong with it. Most briefs arrive asking for the wrong seniority, the wrong title, or a stack combination that does not exist in one person.
RemovesRoles that cannot be filled as written
- 02
Source
Recruiters find candidates
This is the only stage a recruiter owns. Sourcing draws on the parent practices' networks in the platform communities these engineers actually live in, not job-board keyword matching.
RemovesNobody — this stage adds
- 03
Screen
The practitioner screen
A live session against a real production scenario, run by the engineer who wrote the rubric. Not a quiz, not a take-home. The scenario is published on the relevant role page so you can judge the bar before you engage us.
RemovesCandidates who have read about the platform but not run it
- 04
Verify
Work and reference verification
We confirm the specific claims that matter for the role — the migration they led, the incident they owned, the version they ran — rather than collecting generic references.
RemovesClaims that do not survive contact with a former colleague
- 05
Standing
Ongoing standard
Placed engineers stay in contact with the vetting practitioner. When a placement is wrong we say so first and replace, rather than defending it.
RemovesPlacements that stopped working and nobody escalated
A rubric, shown
This is what the screen actually looks like.
Two live examples, taken from the DevOps and ServiceNow ITOM screens. Every role page carries its own. The difference between a senior and a mid answer is rarely knowledge — it is sequencing, and whether they ask the right question before they touch anything.
Scenario put to the candidate
A RabbitMQ quorum queue loses its leader during a rolling Kubernetes node upgrade. Publishers keep accepting writes, consumer lag climbs, and the on-call dashboard shows the cluster as healthy. Walk me through the first ten minutes.
A senior answer contains
- Asks what the quorum size and the `cluster_partition_handling` setting are before touching anything.
- Separates the two failures: leadership re-election is expected, the healthy dashboard is the actual bug.
- Checks whether the pod disruption budget allowed more than one broker down at once.
- Knows that publisher confirms and consumer acks answer 'did we lose messages', and says so rather than guessing.
Where a mid answer stops
- Restarts the pod first and reasons afterwards.
- Treats consumer lag as the incident rather than a symptom of it.
- Cannot say what the cluster does when a quorum is lost versus merely reduced.
Scenario put to the candidate
A client's CMDB reports 94 percent completeness and the executive service dashboard is still wrong. Where do you look, and what do you tell the sponsor?
A senior answer contains
- Separates the three CMDB health scores immediately: completeness is not correctness and neither is compliance.
- Goes to relationships rather than CI count — a populated CMDB with no service relationships produces exactly this.
- Checks the identification and reconciliation rules for duplicate CIs before blaming Discovery.
- Tells the sponsor the dashboard is a data problem, not a reporting problem, and does not soften it.
Where a mid answer stops
- Reports the 94 percent as good news and looks at the dashboard configuration.
- Cannot name the three health dimensions or say which one the number refers to.
- Proposes re-running Discovery without asking what populated the CIs in the first place.
More screens: devops engineers · data engineers · site reliability engineers · servicenow itom developers · enterprise architects
Who vets
Each screen has a named owner.
A screen nobody signs is a checklist. Each of these engineers owns the rubric for their domain and personally runs the session.
Messaging practice lead
AceMQRuns production RabbitMQ and Kafka estates for regulated enterprises, including quorum-queue migrations and Tanzu/Broadcom licensing transitions.
Data platform lead
Duczer EastCloudera CDP upgrades and private-AI builds inside banks; owns the support-matrix and lifecycle practice.
Platform engineering lead
Ace8Kubernetes and DevSecOps delivery across EKS, AKS, GKE and OpenShift; incident response with a 15-minute commitment.
ServiceNow ITOM lead
Duczer EastDiscovery, Service Mapping and CMDB remediation on enterprise estates; service-oriented CMDB modelling.
For candidates
What this costs you, honestly.
The practitioner screen takes about an hour and it is a real conversation with an engineer who does your job. There is no take-home project and no unpaid work.
If you do not pass, you get told why, specifically, by the person who made the call. Most staffing firms will not do that because they cannot — theirs was a keyword match, so there is nothing to explain.
What we will not do
The limits, stated.
- Send approximate candidates for a technology where our bench is thin. Our ServiceNow practice is ITOM; we say so on the page.
- Publish placement statistics we have not earned. No retention or time-to-fill figures until Conntal has real ones.
- Claim a vetter for a platform where no practitioner has confirmed. Where that is pending, the page says pending.
- Screen for culture fit and pretend it is more reliable than the hiring manager's judgement.
The firms behind the standard
Conntal is the staffing arm of Ace8 and AceMQ.
Conntal is its own company with its own bench and its own commercial terms. What it does not have to build from scratch is the technical judgement: three established consulting practices already run these platforms in production, and their engineers screen every candidate.
Enterprise architecture, DevSecOps, cloud-native, extreme automation, data management and incident response across eight service domains.
Messaging and streaming infrastructure — RabbitMQ, Kafka, Pulsar, IBM MQ, Solace — for regulated enterprises in 26+ countries.
Cloudera, WSO2, ServiceNow ITOM, knowledge graphs and agentic AI for financial services and healthcare.
Figures describing Ace8 and AceMQ (130+ enterprises, 26+ countries) are sourced from acemq.com.
FAQ
What buyers and candidates ask
What is the Conntal Vetting Standard?
It is the requirement that every Conntal candidate is screened by a practitioner at Ace8, AceMQ or Duczer East who runs that technology in production for enterprise clients. Recruiters source; engineers decide. The screen is a real production scenario, and we publish it.
Why publish your screening questions?
Because a rubric you can read is evidence and a claim you cannot check is marketing. Publishing the scenario also filters candidates usefully: the people who are put off by it are the people who would not pass it.
Does publishing the questions not let candidates prepare?
It lets them prepare, which is fine. The scenarios have no lookup-able answer — they are judged on how a candidate sequences an incident, what they ask for first, and whether they separate a symptom from a cause. Preparation does not fake that.
Who are the vetting practitioners?
Senior engineers from the three parent practices, each screening only in their own domain: messaging and streaming, data platforms, platform engineering, and ServiceNow ITOM. Each role page names the engineer who owns that screen.
What happens if a placement does not work out?
The vetting practitioner stays attached to the placement. If the fit is wrong we raise it before you do and replace the engineer rather than defending the original decision.
Do you vet for cultural fit as well as technical depth?
We screen for how someone behaves under production pressure and whether they will say 'I do not know' out loud, which in platform work is the fit that matters. Team and culture judgement stays with you; we do not pretend to assess it better than the hiring manager.
Tell us the role. A practitioner scopes it, not a recruiter.
Send the requirement — or the job description you are not happy with — and an engineer who runs this technology will tell you what is missing before we source anyone.