Skip to content

Instantly share code, notes, and snippets.

@pvcodes
Created August 22, 2025 05:27
Show Gist options
  • Select an option

  • Save pvcodes/618e1d1b27f745074f0493643b2ed1d1 to your computer and use it in GitHub Desktop.

Select an option

Save pvcodes/618e1d1b27f745074f0493643b2ed1d1 to your computer and use it in GitHub Desktop.
WITH osh_ranked AS (
SELECT
customer_id,
ca_credit_turnover_amt AS osh_ca,
eff_start_date,
eff_end_date
FROM (
SELECT
customer_id,
ca_credit_turnover_amt,
eff_start_date,
eff_end_date,
ROW_NUMBER() OVER (
PARTITION BY customer_id
ORDER BY eff_start_date DESC, eff_end_date DESC
) AS rn
FROM bsl_ukc_ngcb_bankdata_ukr
WHERE bus_mth = 202505
) osh_sub
WHERE osh_sub.rn = 1
),
cdh_ranked AS (
SELECT
customer_id,
ca_credit_turnover_amt AS cdh_ca,
eff_start_date,
eff_end_date
FROM (
SELECT
customer_id,
ca_credit_turnover_amt,
eff_start_date,
eff_end_date,
ROW_NUMBER() OVER (
PARTITION BY customer_id
ORDER BY eff_start_date DESC, eff_end_date DESC
) AS rn
FROM bsl_ukc_ngcb_bankdata_ukr_CHG1019913799_bkp
) cdh_sub
WHERE cdh_sub.rn = 1
),
latest_data AS (
SELECT
osh.customer_id,
osh.osh_ca,
cdh.cdh_ca
FROM osh_ranked osh
JOIN cdh_ranked cdh
ON osh.customer_id = cdh.customer_id
WHERE osh.osh_ca != cdh.cdh_ca
)
SELECT
xref.ppeaccountid,
xref.cis_customer_id,
latest_data.osh_ca,
latest_data.cdh_ca
FROM bsl_ukc_ngcb_account_cardholder_xref xref
JOIN latest_data
ON xref.cis_customer_id = latest_data.customer_id;
@pvcodes

pvcodes commented Apr 29, 2026

Copy link
Copy Markdown
Author
IMG_6583

@pvcodes

pvcodes commented May 21, 2026

Copy link
Copy Markdown
Author

PROMPT 1 — Generate the 120 practice questions + 4 projects (main study document)
I'm studying for the Microsoft DP-750 certification - Azure Databricks Associate. Here's the link to the study guide: [Link] And the Microsoft learning plan where you'll get the information from, but don't limit yourself to this, use it as the main resource:
I want you to create:
• 20 questions for Section 1: Set up and configure an Azure Databricks environment (15-20%)
• 1 long project/activity covering everything in Section 1
• 20 questions for Section 2: Protect and control Unity Catalog objects (15-20%)
• 1 long project/activity covering everything in Section 2
• 40 questions for Section 3: Data preparation and processing (30-35%)
• 1 long project/activity covering everything in Section 3
• 40 questions for Section 4: Implementation and maintenance of data pipelines and workloads (30-35%)
• 1 long project/activity covering everything in Section 4
Questions must be Microsoft exam style following the same structure as DP-700, DP-203, etc., but about the DP-750 certification. Questions must come WITHOUT answers — I'll send you my answers later for feedback. They must be in English because the exam is in English. Don't limit yourself in length or complexity. Deliver as a Word document.

PROMPT 2 — Generate the calibrated study plan (after diagnostic questions)
I have X days to study for the DP-750 exam. My exam is on [DATE] at [TIME]. I want you to design an objective-based daily plan to get me as close to 100% probability of passing as possible.
Before designing anything, ask me all the questions you need to calibrate the plan to my actual situation — my workspace access, my current knowledge level on each topic, my available study materials, my daily available hours, and any other context you need. Don't assume I know something just because it was in a previous guide.
The plan should be objective-based, not time-based — each day should have a "DONE WHEN" criteria. Focus on closing my real gaps, not covering everything. Deliver as a Word document.

PROMPT 3 — Generate the reading guide (sub-links to read vs skip)
For each documentation page in the study plan, tell me exactly which sub-links I should read, skim, or skip. I don't have time to read every sub-link in every doc page. Use a color-coded system:
• READ (green): exam-relevant, read fully
• SKIM (yellow): read headers and key points only
• SKIP (red): not on the exam, don't waste time
Base the decisions on the official exam guide bullet points. Deliver as a Word document organized by study day.

PROMPT 4 — Generate Round 2 practice questions
I want you to create 40 new practice questions for Sections 1 and 2 of the DP-750, following the Microsoft exam format used in previous tests. All questions must be different from the previous set. Make sure the correct answers are distributed across A, B, C, and D — not all A or all B. Use this exact exam topic list: [paste the official exam guide topics]. No answers included. Deliver as a Word document in English.

PROMPT 5 — Generate feedback document for wrong/doubted answers
Here are my answers to the practice questions: [list answers]. For every wrong answer, give me:
• My answer and the correct one
• Why I got it wrong (the specific concept I misunderstood)
• The key distinction to memorize
• A specific documentation link to read
For every answer I marked as "doubted," reinforce the correct reasoning so I don't doubt it again.
At the end, include:
• Recurring error patterns across all my attempts
• A quick reference card with the key facts I keep getting wrong
• A priority reading list with only the pages that close my specific gaps (estimated reading time)
Deliver as a Word document.

PROMPT 6 — Full review with topic grouping
Review my feedback on all my practice questions across all sections. Make a list of every question I got wrong. Group them by topic and identify my top 5 weakest topics. For each weak topic, provide the specific documentation to re-read plus a cheat sheet of the key facts. The deliverable should mark when my weak-topic list is down to 3 or fewer topics. Deliver as a Word document.

@pvcodes

pvcodes commented Jun 3, 2026

Copy link
Copy Markdown
Author
SELECT
    c.name AS column_name,
    ep.value AS column_description
FROM sys.columns c
LEFT JOIN sys.extended_properties ep
    ON ep.major_id = c.object_id
    AND ep.minor_id = c.column_id
    AND ep.name = 'MS_Description'
WHERE c.object_id = OBJECT_ID('dbo.Customers');

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment