PCED-30-02 deep review — September 29, 2026

The study guide maps all 40 objectives, grouped9/13/13/5, and expands all16 answers. Complete indexed canonical syllabus and MQC bodies were manually read after direct retrieval and the automated monitor timed out. The existing snapshot was retained; this is not a successful automated hash observation. Same-context AI review leaves human review pending.

Credential and learning findings

The complete exam page identifies30-02 active sinceJuly15,2025, seven-year validity and40questions/60minutesplusNDA/75%. Retired30-01 vouchers already issued may still be redeemed. Each current syllabus item has a maximum of one point; no current official questions or partial-credit rules were accessed. The specific exam FAQ's failed-retake seven-day rule is distinguished from a generic policy footer referring to passed/failed attempts. Only selected public policy sections were reviewed; no account, booking or system test was performed.

The standalone practice kit listsUSD29/two tests/up to10 launches each; the USD95 exam/retake/practice bundle says five. The store directs TestCandidate→Practice while the credential FAQ says Learner. These are visible product/instruction differences, not verified equivalent entitlements. No cart action, purchase, redemption or private question access occurred.

Complete Edube PD101 and Pro product bodies support fiveweeks at about onehour/day, freeCore lessons/moduletests, andUSD49Full/Pro interactive work. Pro lists minimum12-month access after redemption. These are provider estimates/features, not observed completion. VendorPD101 itself was only partly readable. Cisco and the store root returned short application shells; Kaggle failed, so previous duration claims were removed. The complete book landing describes a broader thirdedition updated forpandas2.0/Python3.10; no chapters were reviewed or copied.

There are 40 direct receipts:32 successful responses,7 errors and1 blocked. HTTP success is not evidence of a complete course body. Per-source reading boundaries identify selected Python3.13 sections,11 complete NumPy generated references, partial policy/legal excerpts and unavailable interiors. The generic Python index identifies3.14.7, pinned references3.13.15 and local interpreter3.13.14; no exam patch-version inference or runtime update was made.

Original workbook and observed results

The exact public Python file passes 112 checks normally and112 with optimization, with all warnings promoted to errors and no stderr. Its checks stay active under -O. It uses existingCPython3.13.14 andNumPy2.5.2; selectedRuff0.16.4rules E4,E7,E9,F,W,E501 pass at79characters. Exact commands, timestamps, hashes and full outputs are recorded in the operation evidence. Initial development checks are not substituted for these final exact-source executions.

The original fixture contains17 logicalCSV records, including quoted comma/quote/newline content. Its ledger reconciles8accepted,1duplicate,1missing,2nonfinite,1range,1date,1numeric-text and2conflicting-ID records. Validation precedes deduplication; all conflicting valid records for an ID are excluded, while identical normalized duplicates retain the first. Record numbers are not physical line numbers. Reversal preserves accepted IDs. Raw bytes remain unchanged, with SHA-256 c046bc6af8b9b8fd2d48e704675ffb20783303c2558b0aecbc54fa7cba1a3beb. Missing values are never silently imputed as zero.

The accepted durations10/12/14/16/18/20/22/160 sum to272, mean34, median17. Python and NumPy agree on populationSD47.77028364998475 and sampleSD51.068581339214816. North has count4/mean13/median13; South count4/mean55/median21. The two-population-SD rule flagsR08 while retaining it; a separately labeled seven-record sensitivity gives mean16. Pearsonr with the assigned depth variable is0.6395294793910794, with no causal or population inference. Constant/empty/nonfinite/unequal-pair guards and a zero-linear-correlation nonlinear example make the limits concrete.

The workbook also exercises all named list/string/set/dictionary operations, numeric/type conversion, scope/defaults/None, expectedexceptions, if/elif/else, nestedconditions, for/while/break/continue/else, isolatedseededrandom sampling, explicitCSV/textfiles, exactdateformatting, min-maxconstant policy, Counter, conditionalgroups, NumPydtype/axis/views/copies/sequencegeneration/sorting/unique/finite masks. It shows why manual comma splitting is only a restricted exercise. Temporary files are created under the checked private workspace and removed on completion.

Communication and remaining work

The optional exactPowerShell helper uses the existing Windows .NET chart component to export a1100×650PNG from the computed report without a visible window or installed package. The final chart was AI-inspected: fourvalues13/13/55/21, explicitzero baseline,units/counts/source,meaningful title andreadable labels. A table andalt text accompany it. The five-sentence narrative separates observed fixture results from limitations and a proportionate investigation step; analyst/executive presentation guidance explains how to answer questions from the ledger.

This is a bounded synthetic demonstration, not the separate public-dataset lab, a real service evaluation, a complete accessibility review or human readiness approval. The public-dataset activity, current product/redemption confirmation, inaccessible interiors and human review remain explicit. No authentic exam questions or protected course material were copied.