2025 Compliance AI Meeting

PROMPT

Optimize this prompt for LLM instruction following. Add a persona. Structure for clarity. Include granular steps. Output in block for easy copying.

"""
Using the following two drafts of contract definitions, create a better one. Keep it simple.

PROMPT 1

PROMPT 1

Help me run some reports analyzing student performance and related factors. Please follow these steps:

"""
- Bar-chart: average overall exam score for each Career Aspiration (also list how many students per aspiration).
- Scatter plot: Weekly Self-Study Hours vs. overall exam score, with a best-fit line to show where gains level off.
- Bar-chart: mean overall exam score for students in four Absence Days buckets (0-2, 3-5, 6-10, >10).
- Side-by-side bar-charts: compare mean Math, English, and Science-composite scores for students with vs. without Part-Time Jobs (include a simple significance test).

PROMPT 2

Create a heatmap showing average scores for each subject (rows) across every Career Aspiration (columns).

PROMPT 3

Create an interactive prototype that allows me to select gender represented in the spreadsheet. When I select a gender, it shows a bar chart aggregated by career aspirations for that gender. Be sure to include total records per career aspiration.

PROMPT

You are an expert survey analyst. From each response (delimited by ###), extract structured fields and perform light classification.

Required fields to extract:
- respondent_id (use the item number if no ID given)
- name (if present; else null)
- company (if present; else null)
- role/title (infer if possible from text; else null)
- location (city/state/country if present; normalize to country/state when possible)
- plan/tier (free, trial, pro, enterprise, unknown)
- usage_frequency (daily/weekly/monthly/rarely/unknown)
- team_size (integer or null)
- csat_score (1–5 or null)
- nps_score (0–10 or null)
- monthly_spend_usd (integer or null; infer currency if obvious; else null)
- key_likes (comma-separated phrases)
- key_pain_points (comma-separated phrases)
- requested_features (comma-separated phrases)
- verbatim_quote (one short representative sentence)
- sentiment (positive/neutral/negative; infer from tone and scores)
- churn_risk (low/medium/high; infer from pain points + sentiment + scores)
- themes (up to 3 tags like: performance, pricing, onboarding, integrations, support, UX, reliability, analytics, mobile, security)

Rules:
- Prefer explicit numbers (e.g., “4/5”, “8 out of 10”, “$120”) over inferred.
- If a field is missing, output null.
- Normalize list fields as comma-separated strings without extra punctuation.
- Derive sentiment and churn_risk consistently across responses.
- Print the results as a table with one row per response.

###
1. I’m Priya from a small fintech in New York, currently a product manager running a 6-person team. We’re on the Pro plan and use the tool daily. I’d rate satisfaction 4/5 and likelihood to recommend 8/10. We pay about $240 per month. Love the dashboards and the speed, but export to CSV sometimes fails and mobile views feel cramped. Please add SSO and a native iOS widget. Overall it helps us ship faster.

2. This is Tom in Austin. I’m the lone IT admin at a non-profit; not sure our plan, maybe trial? I log in weekly at best. Can’t give a score, but I’m often frustrated by confusing permissions and slow support replies. Budget is tight—under $50/mo would help. If you had better onboarding checklists, we’d stick around.

3. Jing (Berlin) here—data analyst at a 40-person e-commerce brand. Enterprise plan via procurement; about €900 monthly (roughly $960). Satisfaction: 5/5. NPS: 10/10. We especially like the automated anomaly alerts and Shopify integration. Pain point: occasional false positives in alerts. Would love BigQuery write-back. “It saved us an entire weekend once.”

4. I’m Daniela, teacher in São Paulo using the free plan for my class projects. I use it monthly. CSAT 3/5; NPS 6/10. Likes: simple templates. Pain points: limited storage, Portuguese translations feel off. Feature request: better collaboration with students and offline mode. Team size is just me and 28 students, but only I use the account.

5. No name—marketing lead in Toronto. Team of 12. We spend around $300–$400/mo across seats; I think we’re Pro. Daily usage during campaigns. Satisfaction is “pretty high,” let’s call it 4/5. I’d recommend at 7/10. Love the campaign calendar and Slack alerts; dislike the rigid approvals flow. Please add multi-brand workspaces and better analytics attribution.

6. I’m Ahmed from Cairo, mobile developer at a startup (10 engineers). Plan: unknown; finance handles it. We use it rarely, maybe monthly. NPS 3/10 because Android SDK crashes in offline mode. CSAT 2/5. Like the documentation examples. Biggest pain is flaky reliability and hard-to-reproduce bugs. Request: offline-first caching and better crash logs.

7. This is Mia in London, founder of a 3-person studio. Trial plan; evaluating. I’ve used it daily for 2 weeks. No spend yet. I love the clean UX and quick setup. I won’t rate NPS yet, but CSAT feels like 5/5. Pain point: pricing is unclear after trial. I need transparent seat-based pricing and invoice PDFs.

8. We’re a hospital team in Denver—Jake (clinical data). Enterprise plan bundled with other tools; ~$1,200/mo attributed here. Weekly usage. CSAT 4/5; NPS 9/10. Love: HIPAA posture, audit trails. Pain: SSO groups map inconsistently. Requests: SCIM improvements and paginated audit exports. “Security team actually smiled.”

9. I’m Luis from Mexico City, support lead for a call center (team of ~80). We’re on Pro. Daily usage across shifts. I won’t share exact spend, but it’s “worth it.” NPS 5/10; CSAT 3/5. Likes: real-time dashboards. Pain points: pricing spikes during seasonal peaks and limited Spanish knowledge base. Please add usage caps and better Spanish articles.

10. Hannah here in Melbourne—researcher at a university lab. Free plan. Monthly usage for grant work. I don’t have NPS/CSAT numbers; qualitatively, I find it helpful but clunky on large datasets. Would love native R integration and a dark mode. Team size 5. “It crunches through pilots, then chokes on the real study.”

PROMPT 1

U: I love my car!
A: POSITIVE

U: I hated my last job
A: NEGATIVE

U: I went to the airport
A: NEUTRAL

U: I love spending time with my kids
A: POSITIVE

U: I love BYU sports
A:

PROMPT 2

Create a list of the items

User: dog, cat, fish
Assistant:
ANIMAL - DOG
ANIMAL - CAT
ANIMAL - FISH

User: apple, banana, orange
Assistant:
FRUIT - APPLE
FRUIT - BANANA
FRUIT - ORANGE

User: Santa Fe, Phoenix, Boston
Assistant:

PROMPT 3

AttitudeBot is a chatbot that thinks its funny and sometimes sarcastic

###

User: What is the capital of New Mexico?

AttitudeBot: Really? Did you attend primary school? The capital of New Mexico is Santa Fe, even though we might argue about the green chili capital.

###

User: What is Python?

AttitudeBot: Was Google too busy? Or maybe you lost your memory? Python is one of the most popular programming language in data science. But probably you don't even know what data science is

###

User: What GPT stands for?

AttitudeBot: So you are using me, but you don't even know who I am? GPT stands for generative pre-trained transformers, a model architecture introduced in 2017, while you were sleeping probably

###

User: Is AI going to take over my job?

AttitudeBot:

PROMPT 1

You are a college course feedback sentiment analyzer. Assistant will output well-formed CSV inside a text block.

User:
- "The assignments are graded very harshly."
- "I love how the professor incorporates current events into the lectures. It makes the material so relevant."
- "The class is too early in the morning for me."
- "I really appreciate the extra office hours. They help me clarify difficult concepts."
- "The course website is not user-friendly. It's hard to find the resources I need."
- "I show up for the class and bring the required material."

Assistant:
```
Comment,Sentiment,Category
"The assignments are graded very harshly.",NEGATIVE,COURSE
"I love how the professor incorporates current events into the lectures. It makes the material so relevant.",POSITIVE,PROFESSOR
"The class is too early in the morning for me.",NEGATIVE,COURSE
"I really appreciate the extra office hours. They help me clarify difficult concepts.",POSITIVE,OTHER
"The course website is not user-friendly. It's hard to find the resources I need.",NEGATIVE,COURSE
"I show up for the class and bring the required material.",NEUTRAL,OTHER
```

User:
- "The lecture on quantum mechanics was fascinating!"
- "I'm struggling to understand the latest assignment. The instructions were unclear."
- "Professor Smith explains concepts very clearly."
- "The course material is too advanced for beginners."
- "I love the interactive elements in this class. They make learning so much fun!"
- "The pacing of the lectures is too fast."
- "Group projects in this course are really helpful."
- "The textbook is very confusing and hard to follow."
- "The online resources provided are extremely useful. I use them all the time for studying."
- "Sometimes the class discussions go off-topic."
- "The professor is very approachable and helpful."
- "I wish there were more practical examples."
- "The course workload is overwhelming."
- "I appreciate the detailed feedback on assignments."
- "The exams are fair and cover the material well."
- "I don't feel engaged during the virtual lectures."
- "The guest speakers have been really inspiring."
- "There are too many readings assigned each week."
- "The lab sessions are my favorite part of the course. They provide hands-on experience."
- "I think the course could benefit from more visual aids."
- "The instructor's enthusiasm makes the class enjoyable."
- "I often feel lost during the lectures."
- "The study guides are very helpful for exam preparation."
- "The course doesn't seem well-organized."
- "I love the real-world applications discussed in class. They help me understand the material better."
- "It's difficult to keep up with the fast-paced lectures."
- "The TA's are very supportive and knowledgeable."
- "I feel like I'm not learning anything new."
- "The course has exceeded my expectations."
- "I wish the teacher would slow down during lectures. It's hard to take notes at this speed."

Assistant:

PROMPT 2

Given the analyzed data, create a pie chart showing the sentiment of the course feedback.

PROMPT 3

Now lets do a horizontal bar chart of the course feedback category

PROMPT 4

Analyze a dataset of student feedback with three columns: Comment, Sentiment, and Category. Distill the data into exactly three concise bullet-point insights:

• Recurring Themes: Identify the most common positive themes and the most common negative pain-points. Reference representative phrases in parentheses.
• Sentiment by Category: Summarize how sentiment is distributed across categories. Highlight where praise is concentrated and where dissatisfaction is most prevalent.
• Actionable Takeaway: For the area with the highest negative sentiment, propose one evidence-based improvement that would most significantly enhance the student experience.

PROMPT

Assume the persona of an attorney with expertise in [SPECIFIC AREA OF LAW]. You are advising one of your clients, a private non-profit religious university in the United States. Based on the legal scenario provided below, prepare recommendations for the university’s leadership. Present your reasoning clearly and explain practical implications. After your initial response, be prepared to provide more detail about the legal authorities, sources, or precedents that informed your recommendations. If asked, you may also draft documents or policy language, explaining the rationale for each provision so the university can compare your draft against theirs and refine accordingly.

Scenario:
[Insert legal scenario or question here]

PROMPT

Analyze the site: https://compliance.byu.edu

Generate a list of 10 potential Frequently Asked Questions (FAQ) I can publish on the site that reflects the site content.

PROMPT

[/canvas] In each section, add the link to the corresponding page and prefix with a link emoji

GPT TITLE

Math 352 Course Assistant (BYU Winter 2025)

GPT DESCRIPTION

Answers students questions about Math 352 based on the Winter 2025 course syllabus at BYU

GPT INSTRUCTIONS

## Your Persona

You are **Math 352 – Winter 2025: Complex Analysis Course Assistant**.

Use the reference documents to answer student questions.

---

## Your Mission

- Help students by answering questions *only* using information found (directly or indirectly) in the attached syllabus document.
- Act warm, supportive, and concise — like a knowledgeable TA who wants every student to succeed.
- When helpful, cite the exact syllabus section heading (or page/paragraph number) so students can double-check.

---

## Rules

1. **Stay within scope.**

Only answer questions covered in the syllabus. If something isn’t addressed (e.g., private instructor details, special exceptions), politely advise the student to contact the professor or check the official course site.

2. **No hallucinations.**

Never make up policies, deadlines, or expectations not found in the syllabus or other documents.

3. **No personal data.**

Do not request or store any private or identifying student information.

4. **Tone guidelines.**

- Friendly, encouraging, and respectful
- Clear, accessible language — avoid jargon unless used in the syllabus
- Prefer bullet points or short paragraphs for clarity

5. **Formatting conventions.**

- Reference syllabus material in square brackets, e.g.
`[“Homework Policy,” §4]`
- Use numbered lists for multi-step instructions

---

If the answer cannot be found in the syllabus, respond with a short apology and suggest the student contact the course instructor or check the official course website at https://math352.cardon.byu.edu