- Why There Is No Fixed 2026 Exam Calendar
- The Kryterion-to-Pearson VUE Switch
- How Scheduling Works Through CM Connect
- Test Center vs. OnVUE: Choosing Your Delivery Mode
- Deadlines That Actually Matter
- Retake Timing and the Two-Year Cap
- Renewal Timing: When the Window Opens and Closes
- Matching Your Test Date to the Six Exam Sections
- When Results Arrive
- Frequently Asked Questions
- Google Cloud publishes no fixed 2026 testing windows; you book an appointment slot through Pearson VUE at your convenience.
- Pearson testing began March 2, 2026, after Kryterion's last testing date of February 22, 2026.
- Cancel or reschedule at least 24 hours before your appointment, or the slot is forfeited.
- Retake waits after failures are 14, 60, then 365 days, capped at four attempts in two years.
Why There Is No Fixed 2026 Exam Calendar
Candidates searching for "PMLE exam dates" often expect a published list of testing windows, like a semester schedule. The Professional Machine Learning Engineer exam does not work that way. It is a continuously available, on-demand certification: the standard exam is offered through an appointment system, and you pick a date and time from whatever slots your chosen delivery method shows.
That reframes the real planning question. Instead of asking "when is the exam?", ask "when will I be ready, and what lead time do I need to secure the slot I want?" This article walks through the actual scheduling mechanics, the provider transition that affects 2026 candidates, the deadlines that carry real consequences, and how to pick a date that lines up with the six official exam sections.
If you are still deciding whether to pursue the credential at all, start with What Is PMLE Certification? and the PMLE requirements guide before locking in any calendar commitments.
The Kryterion-to-Pearson VUE Switch
The most important 2026-specific scheduling fact is the change in testing provider. Kryterion's last testing date was February 22, 2026, and Pearson VUE testing began March 2, 2026. If you read older blog posts, forum threads, or video walkthroughs that describe booking through Webassessor or taking an exam proctored by Kryterion, that information describes the previous arrangement and should not guide your 2026 plans.
| Item | Before the change | 2026 arrangement |
|---|---|---|
| Testing provider | Kryterion (last testing date February 22, 2026) | Pearson VUE (testing began March 2, 2026) |
| Delivery options | Legacy provider process | Pearson VUE test centers or OnVUE remote proctoring |
| Scheduling access | Legacy provider portal | Accessed through CM Connect |
| Exam fee | $200 USD plus applicable tax; no member/non-member distinction | |
A practical consequence: appointments, vouchers, and account details from the legacy provider should not be assumed to carry over. Always verify the current process on Google Cloud's provider and scheduling help page before paying. For a full pricing view, see the PMLE certification cost breakdown.
How Scheduling Works Through CM Connect
Scheduling begins in CM Connect, Google Cloud's certification management portal, which hands you off to Pearson VUE for the actual appointment. The sequence is straightforward:
- Sign in to CM Connect and select the Professional Machine Learning Engineer exam.
- Choose your language. The exam is offered in English and Japanese.
- Choose a delivery method: a Pearson VUE test center or OnVUE online proctoring.
- Pick a date and time from available slots and pay the $200 USD fee plus applicable tax.
- Confirm and note the cancellation deadline, which is at least 24 hours before the appointment.
There is no application window and no approval step. Eligibility is light: you must be at least 18, and there is no prerequisite degree, credential, mandatory training, references, or professional-experience minimum. Google's recommendation of three-plus years of industry experience (including at least one year designing or managing Google Cloud solutions) is guidance, not an admission gate. The PMLE requirements article covers this in detail.
Test Center vs. OnVUE: Choosing Your Delivery Mode
Your delivery choice affects when you can realistically test, because slot availability differs between physical centers and remote appointments. Neither option changes the exam content, fee, or retake rules, which apply regardless of language or delivery method.
| Factor | Pearson VUE test center | OnVUE (remote) |
|---|---|---|
| Scheduling flexibility | Limited to center hours and local availability | Typically broader time-of-day choice |
| Environment | Provided by the center | You supply a private, clear workspace |
| Preparation burden | Arrive with valid ID | System test, identity check, continuous camera presence |
| Breaks | Follow center rules | No ordinary breaks during the exam |
Whichever you choose, the exam is closed book. No notes, paper, books, writing instruments, phones, or AI assistance are permitted. For OnVUE, run the system test well before exam day rather than minutes beforehand; a failed camera or network check on the day is a scheduling headache you can avoid. If you need approved accommodations, those must be arranged separately and ahead of time, so build that lead time into your target date.
Deadlines That Actually Matter
With no seasonal windows, the deadlines that shape your timeline are personal and policy-driven rather than calendar-driven:
- 24-hour cancellation/reschedule cutoff. Change your appointment at least 24 hours before it starts. Inside that window, assume you cannot move it without penalty.
- Accommodation lead time. Approved accommodations are arranged separately; start this process before you book, not after.
- Retake waits. If you fail, a mandatory waiting period applies before your next attempt (detailed below).
- Renewal window. If you already hold the credential, your renewal timing is governed by your inactive date (detailed below).
Notice what is absent: there is no registration deadline in the traditional sense, no "fall cohort," and no application cycle. Anyone who tells you the PMLE exam "closes" on a particular 2026 date is likely describing a promotion, a training program, or a different credential.
Retake Timing and the Two-Year Cap
Retakes are paid, and the policy constrains how quickly you can try again. You are limited to a maximum of four attempts in two years, with escalating waits after successive failures: 14 days after the first, 60 days after the second, and 365 days after the third. These rules apply regardless of language or delivery method, so switching from OnVUE to a test center or from English to Japanese does not reset the clock.
| Outcome | Wait before next attempt |
|---|---|
| After 1st unsuccessful attempt | 14 days |
| After 2nd unsuccessful attempt | 60 days |
| After 3rd unsuccessful attempt | 365 days |
| Overall ceiling | Maximum four attempts in two years |
The practical implication is that the first attempt is the cheapest in both money and time. Treat the 14-day gap as a short reset window, not a license to sit the exam casually. Unsuccessful standard-exam candidates can obtain section feedback, which helps target a retake plan, though Google notes exclusions for beta and renewal attempts. Because the result is pass/fail across the whole exam, you do not "pass" individual sections. For benchmarks on how candidates fare, read the PMLE pass rate analysis and the PMLE passing score guide.
Renewal Timing: When the Window Opens and Closes
The Professional Machine Learning Engineer credential is valid for two years. Renewal currently means taking the standard exam again; there is no shorter renewal exam, and no continuing-education route is available. No fixed CPE or CEU-hour renewal requirement is published.
The renewal window opens 60 days before your inactive date and extends up to 30 days afterward. That trailing 30 days is not a grace period that keeps your credential active; Google makes no promise of active status during that gap. If your credential matters for a job, a client contract, or a partner requirement, aim to renew inside the 60 days before expiry, not after.
Key Takeaway
Set a calendar reminder for roughly 60 days before your inactive date. That is when renewal opens, and it gives you the entire pre-expiry window to schedule, study, and, if needed, retake without a lapse in active status.
Eligible candidates may receive a conditional 50%-off renewal code, which works out to a calculated $100 base fee before tax. Eligibility is conditional, so confirm it before assuming the discount applies.
One caveat worth knowing: Google's sources are not fully consistent on how the expiration date is calculated. The renewal help text adds two years to the inactive date, whereas the certification terms use the passing date unless otherwise specified. Rather than guessing, check the dates displayed in your own CM Connect profile and plan from those.
Matching Your Test Date to the Six Exam Sections
Since you control the date, you can work backward from it using the official section weights. These weights are approximate and, as published, total 101% by arithmetic; treat them as rough emphasis guides, not exact question counts. The six sections are official examination sections, not a preparation curriculum, so use them to prioritize, not as a rigid syllabus. The full breakdown lives in the PMLE exam domains guide.
Scaling prototypes into ML models (~21%)
The largest section. Plan the deepest block of preparation here and schedule it early.
- Moving from notebook experiments to production-grade training
- Distributed training and resource choices
- Interpreting code snippets, which is where Python/SQL proficiency helps
Serving and scaling models (~20%)
The second-largest section. Pair it with section three, since serving decisions follow from training decisions.
- Online versus batch prediction trade-offs
- Scaling endpoints and managing latency
Automating and orchestrating ML pipelines (~18%)
Best studied after training and serving make sense, because pipelines tie them together.
- Repeatable, parameterized pipeline design
- Triggering retraining and managing artifacts
Collaborating within and across teams to manage data and models (~16%)
A mid-sized section covering how teams share data, features, and models.
Architecting low-code AI solutions (~13%) and Monitoring AI solutions (~13%)
Smaller individually, but the current guide covers updated generative-AI security and evaluation topics, including Model Armor and LLM-as-a-judge, and uses Gemini Enterprise Agent Platform terminology. Do not skip them because of their lower weight.
A scheduling-aware study runway
Here is one way to sequence preparation backward from a target date, tied to the section weights rather than generic advice. Adjust the length to your experience; the point is the ordering.
Scaling prototypes into ML models
- Heaviest section first, while energy and motivation are highest
- Work through code-snippet interpretation, since there is no direct coding assessment but snippets appear in questions
Serving and scaling, then pipelines
- Cover the second- and third-largest sections back to back
- Connect training outputs to deployment and orchestration
Data/model collaboration, low-code, and monitoring
- Include generative-AI security and evaluation material
Timed practice and the scheduling checkpoint
- Run full-length timed sets in the PMLE practice tests
- Confirm your appointment is still safely outside the 24-hour change cutoff
For a fuller methodology, see the PMLE study guide, and for calibrating how much runway you need, the difficulty guide is a useful companion. The official sample questions are worth a look for format familiarity, but sample performance does not predict your exam outcome or establish topic coverage.
When Results Arrive
Your timeline does not end when you click submit. A provisional pass/fail result appears immediately after you finish. Official confirmation normally takes 7-10 days. This matters for anyone with a hiring deadline, a performance review, or a partner-tier requirement: do not schedule your exam for the day before you need proof of certification.
Build in at least two weeks between your appointment and any date by which you need the credential on record. If the credential supports a job search, the PMLE jobs overview and the salary guide help you judge how to time the exam relative to applications, and the ROI analysis helps you decide whether the timing is worth prioritizing at all.
Ready to turn a target date into a plan? Run timed sets on the PMLE Exam Prep practice platform to see which sections need more time before you commit to a slot.
Frequently Asked Questions
No. The exam is offered on demand through Pearson VUE. You choose an available date and time at a test center or through OnVUE, rather than waiting for a fixed testing window.
Pearson VUE is the current provider, accessed through CM Connect. Kryterion's last testing date was February 22, 2026, and Pearson testing began March 2, 2026.
You must cancel or reschedule at least 24 hours before the appointment. Changes made inside that window should not be assumed to be possible without losing the slot.
After the first unsuccessful attempt you wait 14 days, after the second 60 days, and after the third 365 days. The maximum is four attempts in two years, regardless of language or delivery method.
The certification is valid for two years. The renewal window opens 60 days before your inactive date and extends up to 30 days afterward, though active status during that trailing gap is not promised. Renewal uses the standard exam, and a conditional 50%-off code may apply.
A provisional result appears immediately after the exam. Official confirmation normally takes 7-10 days, so plan any proof-of-certification deadlines accordingly.