Jeff Bullas

Hook: You can create a cinematic poster that looks like a movie-house print — even if you’re not a designer. Focus on mood, composition, and iteration.

Why this works: AI speeds up visual idea generation. You control the story: color, lighting, lens, and typography. The result is a strong starting art file you can refine for print or web.

What you’ll need:

• One clear logline or short film theme (1–2 sentences).
• Reference images or moodboard (2–5 photos for style).
• An AI image generator (example types: text-to-image tools) and a simple editor (free or paid).
• Basic export settings: poster aspect ratio ~2:3, 300 DPI for print, 2000–4000 px on the long side for digital.

Step-by-step:

1. Write your short logline: one sentence capturing theme + emotion.
2. Pick a visual style: noir, pastel, retro sci-fi, documentary, etc.
3. Compose a strong AI prompt (sample below). Run the prompt and generate 8–12 variations.
4. Select 2–3 favorites. Do targeted re-prompts to shift color, crop, or lighting.
5. Download the highest-resolution versions. Use a simple editor to add title, credits, and finish (contrast, grain, dodge/burn).
6. Export final: 300 DPI for print, sRGB 8-bit for web.

Prompt (copy-paste):

“Cinematic movie poster for a short film, dramatic and moody, central figure in silhouette against a stormy city skyline at dusk, cinematic lighting, deep blue and orange color palette, high contrast, film grain, shallow depth of field, 50mm cinematic lens, dramatic rim light, bold negative space for title at bottom, vintage poster typography placeholder — ultra-detailed, photorealistic, poster composition, 2:3 aspect ratio”

Worked example:

• Logline: “A lost violinist finds an unexpected audience in an empty station.”
• Style chosen: moody, film-noir with warm candlelight accents.
• Used the prompt above and changed “city skyline” to “train station interior” and adjusted color to warm ambers + cool shadows.
• Picked the best result, cropped to 2:3, added title in bold serif at bottom, added subtle grain and vignette.

Common mistakes & fixes:

• Too many details in prompt — keep priority elements (mood, subject, color, composition).
• Low resolution outputs — always request highest size and upscale if needed.
• Bad typography — choose contrasting font and ensure legibility over busy areas (use bands or overlays).
• Inconsistent lighting — re-prompt specifying “rim light” or “backlit” to unify light source.

Action plan (quick checklist):

• Do: Start with one clear logline and run 10 variations in the AI tool.
• Do: Save references and note which words changed the output you liked.
• Do-not: Rush typography — test title over the image before finalizing.
• Do-not: Assume first result is final — iterate twice.

Final reminder: Treat the AI output as a creative partner — generate fast, refine slowly. Aim to produce one polished poster in an afternoon: ideate, iterate, finish.

Great question. You’re asking about reliability first — exactly the right focus before you automate any part of investor research.

Here’s the pragmatic truth: AI can reliably extract metrics from investor decks and reports when the numbers are in selectable text or clean tables, you give it a structured schema, and you demand evidence. It struggles with charts, footnotes, and inconsistent definitions unless you set guardrails. Aim for “evidence-backed extraction,” not magic.

What you’ll need

• PDFs or slides (ideally text-based; if scanned, run OCR first).
• An AI model that can follow instructions and return JSON.
• A simple spreadsheet for review and normalization.
• 30–60 minutes for a first pass; less once the workflow is saved.

How reliable is it? What to expect

• Text and tables: typically 85–95% accurate with citations.
• Scanned PDFs without OCR: drops to 50–70% until you OCR.
• Charts/graphics-only metrics: 40–70% unless you manually confirm.
• Definitions vary: ARR vs. revenue, bookings vs. billings — AI needs instructions and must cite sources.

The reliability playbook (step-by-step)

1. Prep the files
   • Combine the deck, financials, and any MD&A into one PDF where possible.
   • If the file is a scan, run OCR. Skipping this is the #1 accuracy killer.
2. Define your schema before extraction
   • Pick the 10–15 metrics you actually use: ARR, MRR, revenue growth, gross margin, CAC, LTV, LTV/CAC, logo churn, net revenue retention, burn, runway, cash, GMV, MAU/WAU, cohort retention, unit economics.
   • Decide expected units (USD, %), periods (FY, Q, trailing 12 months), and as-of dates.
3. Use an evidence-first extraction prompt
   • Tell the AI: don’t guess; cite page and quote; return null if not found.
   • Ask for JSON only. This forces consistency and easy review.
4. Run in passes for higher accuracy
   • Pass 1: Text and tables only.
   • Pass 2: If metrics are missing, allow extraction from charts but flag as “low-confidence.”
   • Pass 3: Compute derived metrics (e.g., runway = cash / monthly burn) with inputs you’ve already verified.
5. Normalize and validate
   • Convert currencies and units; align periods (quarter vs. annual).
   • Add simple checks: margins between -10% and 95%, CAC > 0, LTV/CAC 1–15, NRR 70–160%.
   • When conflicts appear, prefer audited financials over slides and cite both.
6. Human review on exceptions
   • Scan all “null” and “low-confidence” fields first; these are your quick wins.
   • Spot-check any value without a clean citation.

Copy-paste prompt (use as-is)

Extract the following metrics from this investor deck/report. Return JSON only. For each field, include: value, unit, period (Q/FY/TTM), as_of_date, source_page, source_quote, confidence (high/medium/low). Do not guess. If not explicitly found in text or tables, return null and a reason in “note”. If numbers conflict, choose the most recent audited source and list alternatives in “note”. Metrics: ARR, MRR, Revenue, Revenue_Growth_% (YoY), Gross_Margin_%, CAC, LTV, LTV_to_CAC, Logo_Churn_% (annual), Net_Revenue_Retention_% (NRR), Burn_Rate (monthly), Runway_Months, Cash, GMV, MAU, WAU. JSON template: { “ARR”: {“value”: null, “unit”: “USD”, “period”: null, “as_of_date”: null, “source_page”: null, “source_quote”: null, “confidence”: null, “note”: null }, … }

Insider trick

• Evidence-gated extraction: Make the model refuse to fill a field without a direct quote and page number. This single rule boosts precision more than any fancy model tweak.
• Triangulate derived metrics: For runway, require both cash and burn citations; if one is missing, return null instead of guessing.

Quick example (what good output looks like)

• ARR: value 24,000,000; unit USD; period FY2024; as_of_date 2024-12-31; source_page 7; source_quote “FY24 ARR: $24M”; confidence high.
• NRR: value 118; unit %; period FY2024; source_page 9; quote “Net dollar retention: 118%”; confidence medium.
• Runway_Months: value 14; unit months; period current; source_page 18; quote “Cash: $14.7M; Burn: $1.05M/mo;” confidence high; note “Computed: cash/burn.”

Common mistakes and fast fixes

• Mistake: Asking for “key takeaways.” Fix: Provide a field-by-field schema and demand JSON.
• Mistake: No OCR on scans. Fix: OCR first; then re-run extraction.
• Mistake: Accepting values without citations. Fix: Require page and quote per field; reject anything else.
• Mistake: Mixing time periods (Q vs. FY). Fix: Force “period” and “as_of_date” in the output.
• Mistake: Chart-only metrics treated as facts. Fix: Mark as low-confidence or null, and confirm manually.
• Mistake: Single-pass extraction. Fix: Use the two-pass (text/tables → charts) approach.

Action plan (today)

1. Pick one recent deck and one report (10–20 pages).
2. OCR if needed; save as one PDF.
3. Run the copy-paste prompt above with your metric list.
4. Sort outputs by confidence; review nulls and lows first.
5. Normalize units and periods in your spreadsheet.
6. Save the prompt and schema as your house template for the next deal.

Bottom line

AI is reliable for metric extraction when you control the process: schema-first, evidence required, and honest nulls instead of guesses. Expect high accuracy on text and tables, lower on charts, and fast iteration after your first template is in place. Start small, demand citations, and you’ll get dependable, investment-grade summaries in minutes.

You’re right to want a simple, non-overwhelming start. Let’s build a tiny “second brain” you can set up in minutes, then grow only if it earns its keep.

Try this now (under 5 minutes):

• Create one folder called “Second Brain”.
• Inside it, create two notes: “Inbox” and “Today”. Pin them.
• Open your favorite AI chat. Paste the prompt below and keep that chat open as your helper.

What you’ll need

• A notes app you already use (Apple Notes, Google Keep, Notion, Evernote—any is fine).
• One AI chat (ChatGPT or similar).
• Your phone or computer. That’s it.

The big idea (so you don’t get overwhelmed)

• The Rule of One: one place to capture (Inbox), one place to act (Today), one AI prompt to clarify.
• Focus on two moves only: capture → clarify. Organizing comes later, in tiny passes.
• Use a small, fixed tag list to avoid chaos.

Copy-paste AI prompt (your always-on clarifier)

“Act as my Second Brain Clarifier. When I paste messy notes, do the following and keep it concise:

• 1) Title (max 8 words, action-first)
• 2) 3-sentence summary (plain English)
• 3) Next 1 step (starts with a verb; due date if mentioned)
• 4) Tags: choose up to 3 from ONLY this list: Learning, Work, Money, Health, Family, Home, Travel
• 5) Calendar suggestion (if any date is implied; use YYYY-MM-DD)
• 6) Key references or quotes (optional)

Format with short headings and bullets. If info is missing, say ‘None’—don’t invent. Always keep it brief.”

Step-by-step: your first day

1. Capture: Throughout the day, dump thoughts, links, photos, voice-to-text into your “Inbox” note. Don’t tidy yet.
2. Clarify with AI: At a break, paste one messy note into your AI chat. It will return a clean title, summary, next step, and tags.
3. Act: Move any immediate task to “Today” (just the single next step). If it belongs to an ongoing project, paste the AI output into a new note titled “PROJECT – [Name]”.
4. File lightly: Add the 1–3 tags AI suggested at the bottom of your note. Done. No elaborate folders.

Example

Messy input you paste into AI: “Article about strength training for 40+, consider 2x/week. Doctor said watch shoulder. Maybe buy adjustable dumbbells. Goal: feel stronger by spring. Amy can join?”

What you should expect back:

• Title: Start 2x/Week Strength Plan
• Summary: Begin a simple strength routine tailored for 40+. Be mindful of shoulder; start light and focus on form. Consider equipment and an accountability partner.
• Next 1 step: Book a 20-minute plan setup for Saturday morning (YYYY-MM-DD).
• Tags: Health, Family
• Calendar: Saturday (YYYY-MM-DD) 09:30 – Create 8-week strength plan.
• References: Adjustable dumbbells; form cues for shoulder safety.

Insider trick: Name notes for search. Start titles with a verb and, if time-bound, add the date at the end. Example: “DECIDE: New laptop – 2025-02”. You’ll find things faster later.

A simple template you can reuse (paste into any project note)

• Outcome: What “done” looks like (1 sentence)
• Why it matters: 1–2 lines
• Milestones: 3–5 checkpoints
• Next 1 step: Single action with a date
• Notes: Quick bullets only
• Tags: Pick up to 3 from your fixed list

Common mistakes and quick fixes

• Too many tools: Stick to one notes app and one AI chat for 30 days. Re-evaluate after.
• Tag explosion: Use only the 7 tags listed in the prompt. If a new tag is tempting, map it to an existing one.
• No daily touch: Spend 5 minutes each evening—process 3 Inbox items, tops. Small wins build momentum.
• Over-detailing: Stop when you have a clear next step. Complexity kills consistency.
• Privacy: Don’t paste sensitive info into AI. For private items, summarize in your own words before asking AI.

What success looks like in week one

• Your Inbox is emptied once a day (even if you only process 3 items).
• Each active project has a single, clear “Next 1 step”.
• You can find any note in under 30 seconds using titles and tags.

Seven-day action plan

1. Day 1: Set up folder, two notes, paste the prompt into AI. Process 3 Inbox items.
2. Day 2–3: Capture freely. Each evening, clarify 3 items with AI. Move only next steps to “Today”.
3. Day 4: Create 1–2 Project notes using the template.
4. Day 5–6: Practice the naming trick and calendar suggestions. Keep tags strict.
5. Day 7: Review: What felt useful? Remove any step you didn’t use. Your system should feel lighter next week.

Final nudge: A second brain isn’t a giant system—it’s a tiny habit. One Inbox. One Today note. One reliable AI prompt. Start small, and let usefulness—not perfection—decide what you add next.

On your side,

Jeff

Great question — focusing on estimating timelines and resource needs is exactly the right place to start. Here’s a quick win you can try in under 5 minutes and a clear, practical method to scale it.

Quick win (5 minutes): Paste a simple task list into an AI model and ask for optimistic/realistic/pessimistic time estimates plus recommended roles. You’ll get a usable ballpark to start a conversation with your team.

What you’ll need

• A short task list or project outline (5–15 items).
• Basic team roles (e.g., PM, designer, developer, QA).
• Any historical data or past project lengths (optional but helpful).

Step-by-step

1. Write a plain-task list. Keep each task one line (e.g., “Design homepage mockups”).
2. Pick a complexity scale: Low / Medium / High for each task.
3. Use the AI prompt below (copy-paste) with your tasks and roles.
4. Review the AI’s estimates. Ask it to show assumptions and dependencies.
5. Adjust with your team’s input and convert into a timeline (Gantt-lite: sequenced list) and a resource plan (who does what, when).

Copy-paste AI prompt

“I have the following project tasks: [paste your task list]. For each task, give three duration estimates: optimistic, realistic, pessimistic (in days). List required roles, effort per role (hours), key dependencies, and the main assumption for each estimate. Summarize the overall realistic timeline and required FTEs week-by-week.”

Example

• Task: Design homepage mockups — Low complexity. AI returns: optimistic 2 days, realistic 4 days, pessimistic 7 days; roles: designer (24h), PM (4h); depends on content availability.
• Combine estimates across tasks, add dependency order (design before development), and you get a 4–6 week timeline with 0.6 FTE designer during weeks 1–2.

Common mistakes & fixes

• Relying solely on AI numbers — fix: always validate with the team and past data.
• Missing dependencies — fix: ask AI to list them explicitly and map task order.
• Using too-fine granularity — fix: group very small tasks so estimates are meaningful.

7-day action plan

1. Day 1: Draft task list and roles.
2. Day 2: Run the AI prompt and collect estimates.
3. Day 3: Review assumptions and dependencies with lead team.
4. Day 4: Adjust estimates using historical data.
5. Day 5: Produce a simple timeline and resource allocation.
6. Day 6: Present to stakeholders for buy-in.
7. Day 7: Schedule regular check-ins and re-estimate as you learn.

Final reminder

AI gives fast, directionally useful estimates. Use it to accelerate planning, not to replace judgement. Start small, validate quickly, and iterate — that’s how safe, useful forecasting becomes part of your routine.

Good call on wanting to avoid overwhelm — that’s the smartest place to begin. Keeping things small and useful wins every time.

Here’s a simple, practical plan to build a “second brain” with AI without getting bogged down. The goal is useful notes you can act on, not a perfect system.

What you’ll need

• A single place for notes (a simple notes app, a folder of text files, or a tool you already use).
• An AI assistant you can paste text into (chat-based or a summarizer tool).
• 10–20 minutes twice a week to process new items and 15 minutes weekly to review.
• A tiny tagging/naming rule: 3–4 consistent tags or a folder name.

Step-by-step — do this in your first 30 minutes

1. Pick one place for all notes. Move one folder or pick one app — stop splitting.
2. Capture anything important immediately: article link, idea, meeting notes.
3. Use AI to summarize: paste the text or link into the assistant and ask for a short summary, 3 key points, and suggested tags.
4. Save the AI output into your note with a date and 1–2 tags.
5. Once a week, open 10 recent notes and convert each into one action or archive it.
6. Repeat. Small, consistent processing builds the habit and the value.

Quick copy-paste AI prompt (use as-is)

“You are my note-taking assistant. Summarize the following text in one sentence, list the three most important points, suggest two practical action steps I could take based on it, and give three short tags (single words) for organizing the note. Text: [paste article or notes here]”

Worked example

Say you read a 800-word article about building habits. Paste it into the prompt above. AI returns: one-sentence summary, 3 key points (make tiny habit, track 2x week, tie to existing routine), two actions (start a 2-minute habit tonight; set a calendar reminder), and tags like “habits”, “productivity”, “routine”. Save that output in one note titled “Habits — 2025-11-22” and tag the note.

Common mistakes & fixes

• Do not chase the perfect tool. Fix: choose one app and stick with it for 30 days.
• Do not save everything without processing. Fix: apply the AI prompt and create one action or archive.
• Do keep tags tiny and consistent. Do review weekly for relevance.

7-day action plan (quick wins)

1. Day 1: Pick your note place and set up 3 tags (10–15 min).
2. Days 2–5: Process 1–3 items/day with the AI prompt (5–10 min each).
3. Day 7: Review 10 notes, turn each into an action or archive (15 min).

Small steps, weekly review, and one reliable prompt will give you a functioning second brain in days — not months. Start today: capture one thing, run the prompt, save and tag it.

You’re asking the right question: the bottleneck isn’t taking notes—it’s turning them into action. Routing your notes straight into Todoist or Notion is a fast win that clears mental clutter and boosts follow-through.

Quick checklist: do / do not

• Do use a consistent task schema (task, owner, due date, priority, tags).
• Do ask AI to output in the exact format your tool expects (Todoist Quick Add lines or a Notion-ready table).
• Do convert relative dates to real dates (e.g., “next Friday” → 2025-02-14).
• Do keep each task short and action-focused (verb first).
• Don’t dump everything—filter for actions, not ideas or decisions.
• Don’t overstuff fields; if a detail isn’t in the note, leave it blank.
• Don’t rely on memory—always review the AI’s first pass before bulk-adding.

What you’ll need

• A note (meeting notes, voice transcript, brainstorm).
• Access to an AI chat tool.
• Todoist account or a Notion database for tasks (with properties like Title, Due date, Priority, Tags, Notes).

Fast path 1: Manual, 2-minute convert (copy/paste)

1. Copy your raw notes.
2. Paste into your AI with one of the prompts below.
3. Review the output and paste into Todoist or Notion.

Robust prompts you can copy-paste

• For Todoist Quick Add (makes one task per line you can paste individually into Quick Add):“From the notes below, extract actionable tasks only. Output one task per line formatted for Todoist Quick Add as:Task name p2 #Project @tag due YYYY-MM-DD HH:MMRules: Keep task names under 80 chars. Convert relative dates to real dates (YYYY-MM-DD, 24h). Only include fields present in the text (omit unknown project/tags/times). Default priority to p2 if not stated. Limit to 10 tasks. Output just the lines, no extra text.

  NOTES:
  [Paste your notes here]”

• For Notion database (CSV table):“Extract actionable tasks from the notes into a CSV with headers:Task,Due Date,Priority,Tags,NotesRules: Dates in YYYY-MM-DD. Priority as 1–4 (default 2 if unstated). Tags comma-separated. Notes max 120 chars. Only include true action items. Limit to 20 items. Output only the CSV.

  NOTES:
  [Paste your notes here]”

Fast path 2: Native tools inside each app

• Todoist
  • Use Quick Add with natural language: “Call sponsor Tuesday 14:00 p1 #Launch @calls”.
  • Insider tip: Ask AI to produce Quick Add–ready lines. Then paste each line into Quick Add for instant parsing (dates, priority, labels).
• Notion
  • Put your notes in a Notion page. Highlight the text and use Notion AI to extract “Action items.”
  • Turn the result into a simple table (Task, Due date, Priority, Tags, Notes) and move rows into your Tasks database.
  • Expectation: you’ll still eyeball dates and priorities once as quality control.

Automation path (no-code)

1. Pick a trigger (e.g., new Google Doc in a “To-Process” folder or an email you forward with subject “Notes → Tasks”).
2. Add an AI step to extract tasks to structured data (JSON or CSV) using the schema: title, due_date (YYYY-MM-DD), priority (1–4), tags, notes.
3. Create tasks via:
   • Todoist: Create Task action mapping title, due_date/time, priority, labels, description.
   • Notion: Create Database Item mapping Title, Due date, Priority, Tags, Notes.
4. Test with a small note, then turn it on.

Worked example

Raw notes:

• Website tweaks: update pricing page by next Friday. Ask Maria for new FAQ copy. Fix broken checkout error #1245. Reach out to sponsor re Q2 budget Tue 2pm. Post-launch review end of month.

AI output for Todoist Quick Add (paste each line into Quick Add):

• Update pricing page p1 #Website due 2025-02-14
• Ask Maria for new FAQ copy p2 #Website
• Fix checkout error #1245 p1 #Website
• Reach out to sponsor about Q2 budget p2 #Partnerships due 2025-02-11 14:00
• Plan post-launch review p2 #Website due 2025-02-28

AI output for Notion CSV (import into your tasks database or paste into a Notion table):

• Task,Due Date,Priority,Tags,Notes
• Update pricing page,2025-02-14,1,Website,Page layout and pricing tiers
• Ask Maria for new FAQ copy,,2,Website,Request draft and review
• Fix checkout error #1245,,1,Website,Bug triage and hotfix
• Reach out to sponsor about Q2 budget,2025-02-11,2,Partnerships,Confirm call at 14:00
• Plan post-launch review,2025-02-28,2,Website,Collect metrics

Common mistakes and quick fixes

• Vague tasks: “Pricing page” → Fix: “Update pricing page” (verb + object).
• Missing dates: AI skipped dates it couldn’t infer → Add one-line context to your notes: “Deadlines in GMT on business days.”
• Too many tasks: Cap to top 10; create a separate “Backlog” tag for the rest.
• Mismatched tags/projects: Keep a short, fixed list and include it in the prompt so AI maps correctly.

Insider trick: Train a micro-syntax in your notes so AI nails it every time. Example: start action lines with “- [ ]”, add dates in brackets, and priority as (p1–p4). E.g., “- [ ] Update pricing page [2025-02-14] (p1) #Website”. Your prompt can then say, “Only convert lines matching that pattern.” It reduces false positives and speeds review.

15-minute action plan

1. Decide your schema: Title, Due date, Priority (1–4), Tags, Notes.
2. Copy one of the prompts above into your AI tool.
3. Run it on a real note; review the output for 60 seconds.
4. Send to Todoist via Quick Add or paste/import into your Notion tasks database.
5. Optional: set up a simple automation with an AI extraction step for future notes.

What to expect

• First run: 5–10 minutes, with minor edits.
• After you standardize your schema and prompt: under 2 minutes per note.
• Quality improves fast as you keep prompts and tags consistent.

Starting simple beats waiting for perfect. Get one note flowing into tasks today; refine the prompt and fields next time.

Hook: You can turn messy notes into tidy, actionable tasks in 15–30 minutes using AI and simple automations. Let’s make your notes work for you — not the other way round.

Why this matters: If you capture ideas in notes but never convert them into tasks, they don’t get done. AI can read, extract, prioritize and format tasks for Todoist or Notion, saving time and mental overhead.

What you’ll need

• Accounts: Todoist or Notion (or both)
• An automation tool: Zapier, Make, or a simple script using the Notion/Todoist API
• An AI: ChatGPT or another LLM (via web or API) to parse and structure notes
• Sample notes you want converted

Step-by-step: build a simple workflow

1. Decide the source: pick where your notes live (phone notes app, email, Notion page).
2. Create a trigger in your automation tool: new note or new page created.
3. Send the note text to the AI with a prompt that asks for tasks, priorities, due dates and tags.
4. Parse the AI response into structured fields (task title, description, due date, priority, project/tag).
5. Create the task in Todoist or append a row/page in Notion with those fields.
6. Test with 3–5 real notes, tweak the prompt and mapping until results are reliable.

Copy-paste AI prompt (use as-is)

“You are an expert assistant. Read the note below and extract all actionable items. For each action, provide: 1) a short task title (5–8 words), 2) an optional due date in YYYY-MM-DD or ‘none’, 3) priority (low/medium/high), 4) tags (comma separated), and 5) one-sentence context/description. Return the result as a JSON array of objects with keys: title, due_date, priority, tags, description. Note: prefer due dates this week if clearly urgent; if not specified, use ‘none’. Note: do not include non-actionable ideas.”

Example

Note: “Call Sarah about the Q3 budget; prepare 3 slides; follow up on invoice #204; reschedule dentist.”

AI output (example):

• title: “Call Sarah about Q3 budget” — due_date: “2025-11-25” — priority: high — tags: finance,call — description: “Discuss Q3 budget adjustments and decisions.”
• title: “Prepare 3 Q3 slides” — due_date: “2025-11-27” — priority: medium — tags: slides,work — description: “Create three slides summarizing key budget points.”
• title: “Follow up invoice #204” — due_date: “none” — priority: medium — tags: finances — description: “Confirm payment status and next steps for invoice 204.”
• title: “Reschedule dentist” — due_date: “none” — priority: low — tags: personal — description: “Find a new appointment time in the next two weeks.”

Common mistakes & fixes

• Too many false tasks: tighten the prompt rules to ignore non-actionable notes.
• Wrong dates: have AI prefer “none” unless explicit, then let you set dates in Todoist/Notion.
• Formatting errors: validate AI JSON in your automation before creating tasks.

Quick action plan (next 30 minutes)

1. Pick one source (phone notes or Notion).
2. Set up a trigger in Zapier or Make for new notes.
3. Use the prompt above to extract tasks and map fields to Todoist/Notion.
4. Run 5 test notes and adjust priority/date rules.

Closing reminder: Start small, review AI results, and improve the prompt. Automate the routine; keep the decisions.

Quick win: Copy the AI prompt below, paste it into ChatGPT (or your AI tool), add the job title and three strengths from your resume — you’ll get a tailored first paragraph in under a minute.

Why this matters: hiring managers spend seconds scanning. A crisp, targeted cover letter opens the door. AI helps you write faster while keeping control of the message.

What you’ll need

• The job title and company name.
• Three resume highlights (one-line each — achievements, numbers, or skills).
• Your preferred tone (friendly, professional, confident).
• An AI tool (ChatGPT, Bard, MS Copilot, etc.).

Step-by-step (10 minutes)

1. Open your AI tool and paste the AI prompt below. Replace placeholders with the job title, company, and your three highlights.
2. Ask the AI to produce 3 short variations (intro paragraph + 2-sentence value proposition) in your chosen tone.
3. Pick the best variation. Edit one sentence to add a personal detail (why this company or role matters to you).
4. Ask the AI to expand your chosen paragraph into a 3-paragraph cover letter (intro, 1 concrete achievement, closing call-to-action).
5. Run a quick grammar check, read aloud, and save as PDF or DOCX with a clear filename: LastName_CoverLetter_Company.pdf.

AI prompt (copy-paste and replace brackets)

Write 3 short cover letter openers for the role of [Job Title] at [Company]. Each opener should be 1–2 sentences and include: 1) one named achievement from my resume, 2) the skill I’ll bring, and 3) a sentence tying it to [Company]’s goals. My resume highlights: [Highlight 1], [Highlight 2], [Highlight 3]. Tone: [friendly/professional/confident]. After the 3 openers, provide one full 3-paragraph cover letter using the best opener. Keep the full letter ~180–220 words, use active verbs, and end with a concise call-to-action.

Example

Job: Senior Marketing Manager at GreenTech. Highlights: increased email open rates by 40% with segmentation; led a cross-channel campaign driving $250K in revenue; managed a team of 4. One AI-generated opener might be: “As a marketer who boosted email open rates by 40% through segmentation, I can help GreenTech grow engagement and revenue by making your customer communications more relevant.”

Mistakes and fixes

• Generic language — Fix: add one specific metric or outcome from your resume.
• Too long — Fix: keep sentences short; aim for 180–220 words total.
• Misses culture fit — Fix: mention one company value or recent initiative and tie your skill to it.

Action plan (next 30 minutes)

• Try the prompt now and get 3 openers.
• Customize one sentence with a personal reason you want the role.
• Send the final letter to a friend or mentor for a quick read.

Reminder: AI speeds the writing but you control the content. Add your facts, tweak the tone, and you’ll have a compelling cover letter ready in minutes.