"The best products are built by people who deeply understand the humans they're building for."
This document defines the primary user personas for LoopKit. Every feature, command, and interaction pattern should be evaluated against these personas.
| Attribute | Detail |
|---|---|
| Age | 28-35 |
| Background | Former senior engineer at FAANG or high-growth startup |
| Location | Remote — Bali, Lisbon, or Austin |
| Revenue | $0-5K MRR (pre-product-market-fit) |
| Tech Stack | React/Node, Python, or mobile (Swift/Kotlin) |
| Twitter Bio | "Building in public. Previously @company. DM open." |
Sarah left her $300K engineering job to build her own SaaS. She has deep technical skills but has never been a founder. She's read The Lean Startup and follows Pieter Levels and Marc Lou on Twitter. She has 3-4 "ideas" in her Notes app but is paralyzed by fear of building the wrong thing.
She started building a project 6 months ago, got 80% done, then realized nobody wanted it. Now she's terrified of making the same mistake. She needs a system that forces her to validate before building and ships something every week.
- Validate before building — Never again spend 6 months on something nobody wants
- Ship consistently — Something public every single week
- Build an audience — Share her journey and attract early adopters
- Reach ramen profitability — $5K MRR to cover her lifestyle
| Pain | Current Behavior | Emotional State |
|---|---|---|
| Analysis paralysis | Spends weeks "researching" without coding | Anxious, stuck |
| No accountability | Starts projects, abandons them at 80% | Frustrated, self-critical |
| Building in vacuum | No feedback loop with real users | Isolated, guessing |
| Scattered tools | Notion for docs, Trello for tasks, Twitter for updates | Overwhelmed, fragmented |
| Sunday scaries | Feels guilty about what she didn't ship | Defeated, unmotivated |
Week 0: loopkit init
- Has an idea for "a Notion alternative for developers"
- Runs
loopkit init notion-devand answers the 5 questions - AI reveals her ICP is too broad (score: 4/10)
- Narrows to "Notion alternative for technical writers at API companies"
- Gets a score of 8/10 with a clear validation plan
- Feels relief — she has direction now
Week 1-2: loopkit track
- Creates tasks.md with her weekly plan
- Commits with
[#1] Set up landing page— task auto-closes - Sees shipping score: 60% (3/5 tasks done)
- Gets stale task alert for "Research competitors" — snoozes it
Week 3: loopkit ship
- Ships the landing page + waitlist
- AI generates HN title, Twitter thread, and IH post
- Uses the Twitter draft verbatim, gets 50 signups
- Saves ship log — her first public launch
Week 4: loopkit pulse
- Runs
loopkit pulse --share— gets feedback URL - Shares in 3 relevant Discord communities
- Gets 12 responses over the week
- AI clusters: "Fix now: can't tell what the product does from landing page"
- Tags it to sprint — Week 5 priority is clear
Week 5: loopkit loop
- Sunday ritual: sees 80% shipping score, 12 pulse responses
- AI synthesis: "The One Thing: Rewrite hero copy with a clear before/after"
- Accepts recommendation
- BIP post auto-generated: "Week 5: 50 signups, 1 painful lesson about clarity..."
- Posts it, gets engagement, feels momentum
- Time from idea to first user conversation: < 2 weeks
- Shipping score consistency: > 70% weekly
- Loop override rate: < 30% (she trusts the system)
- Weeks until first paying customer: < 8
| Attribute | Detail |
|---|---|
| Age | 24-32 |
| Background | Self-taught developer, dropped out of college or never went |
| Location | Southeast Asia or Eastern Europe (geo-arbitrage) |
| Revenue | $2K-20K MRR across multiple projects |
| Tech Stack | No-code + light code (Next.js, Supabase, Stripe) |
| Twitter Bio | "7 products. 2 hits. Building in public." |
Marcus has already shipped 5-7 projects. Two made money, the rest died. He moves fast — sometimes too fast. He'll build an MVP in a weekend, launch on Product Hunt, and move on if it doesn't get traction in 2 weeks. He's obsessed with shipping velocity but sometimes skips validation.
He doesn't need hand-holding. He needs a system that keeps him honest about validation, helps him manage multiple projects, and automates the boring parts (writing launch copy, collecting feedback, weekly retrospectives).
- Ship more projects faster — 1 new project every 4-6 weeks
- Kill projects faster — Kill failures in week 2, not month 6
- Reuse what works — Build a playbook from past projects
- Grow audience — Build in public consistently
| Pain | Current Behavior | Emotional State |
|---|---|---|
| Skips validation | Builds first, validates later (or never) | Overconfident, then disappointed |
| Managing multiple projects | Different repos, different task systems | Scattered, loses context |
| Writing launch copy | Stares at blank screen for 30 mins | Bored, delays launch |
| No structured reflection | Forgets lessons from past projects | Repeating mistakes |
| Feedback scattered | DMs, emails, comments across platforms | Overwhelmed |
Project A (Active): AI image generator for designers — Week 3 Project B (On hold): Chrome extension for Twitter — Paused at Week 2
Monday Morning: loopkit track
- Switches to Project A:
loopkit track --project ai-image-gen - Sees 4 open tasks, 2 done from last week
- Adds new task:
--add "Implement style transfer endpoint" - Shipping score: 50% — needs to catch up
Wednesday: Context Switch
- Gets an idea for Project C while showering
- Runs
loopkit init chrome-writer --template saas - 10 minutes later: has a scored brief, validation plan, and AI-personalized tasks.md
- Decides to park it for now (score: 6/10, needs more ICP research)
- Back to Project A
Friday: loopkit ship
- Ships style transfer feature
- Uses AI-generated Twitter thread — edits it slightly
- Launches on Twitter + IH simultaneously
- Gets 200 visitors, 20 signups
Sunday: loopkit loop
- 3-week streak active
- AI synthesis: "Kill or pivot? Traction is below threshold for Week 3"
- Marcus overrides: "I want to give it 2 more weeks" — records reason
- Override rate now 60% — system warns him gently
Week 4: loopkit pulse
- Runs
loopkit pulse --sharefor Project A - Embeds widget on landing page
- Gets 8 responses automatically
- AI clusters: "Validate later: want API access" — signals to build API tier
- Projects started per quarter: 3-4
- Projects killed before Month 2: 70%
- Time from ship to feedback analysis: < 10 minutes
- Audience growth from BIP posts: +10% followers/month
| Attribute | Detail |
|---|---|
| Age | 32-40 |
| Background | Product manager, consultant, or domain expert (non-engineer or light technical) |
| Location | Major city — NYC, London, SF |
| Revenue | $0 (pre-launch, may have raised small pre-seed) |
| Tech Stack | Hires freelancers, uses no-code tools |
| LinkedIn Bio | "Founder @ [stealth]. Ex-[bigco]. Looking for technical co-founders." |
Alex has deep industry expertise (e.g., 10 years in logistics, healthcare, or fintech) and a painful problem they've experienced firsthand. They know the problem is real but don't know how to validate it systematically or what the MVP should look like.
They're considering raising a pre-seed round but investors keep asking "Have you validated this with users?" and they don't have a good answer. They need a framework that forces rigor without requiring technical expertise.
- Validate the problem — Talk to 20+ potential users before building
- De-risk the idea — Get evidence to raise money or hire confidently
- Learn to ship — Understand the founder mindset, not just the manager mindset
- Build a network — Find early adopters and potential co-founders
| Pain | Current Behavior | Emotional State |
|---|---|---|
| Don't know what to build first | Writes long PRDs nobody reads | Confused, overthinking |
| Can't code the MVP | Depends on freelancers who need direction | Powerless, frustrated |
| No validation framework | Asks friends if it's a good idea | Biased, insecure |
| Imposter syndrome | Compares self to technical founders | Anxious, doubtful |
| Investor rejection | Pitching without traction data | Demoralized |
Week 0: loopkit init
- Has an idea: "AI-powered freight matching for small importers"
- Struggles with "Who is your ICP?" — types "small businesses"
- Gets soft warning: "Try to be more specific"
- Refines: "Import/export brokers doing <$5M/year who use WhatsApp to coordinate shipments"
- AI scores ICP 9/10 — the specificity pays off
- Riskiest assumption identified: "Brokers will trust AI over their existing WhatsApp network"
- Validation plan: "Interview 10 brokers, ask about coordination tools and trust"
- Feels empowered — has a clear plan for the first time
Week 1-4: loopkit track
- Tasks are non-technical: "Interview broker #1", "Validate trust assumption", "Draft landing page copy"
- Shipping score is based on validation milestones, not code commits
- Uses
--addto capture insights from each interview
Week 5: loopkit ship
- Ships a landing page + Calendly link (built with Webflow)
- AI generates Indie Hackers post about the validation journey
- Gets 5 calls booked from the post
Week 6: loopkit pulse
- Collects feedback from the 5 calls
--add "Broker #3: 'I don't trust algorithms with my client relationships'"- AI clusters this as "Validate later: trust concerns" — pattern emerges
Week 7: loopkit loop
- AI synthesis: "The One Thing: Build a human-in-the-loop feature that lets brokers review matches before sending"
- This becomes the core product insight
- Alex now has validation data to show investors
- User interviews completed in first month: 15+
- Validation insights that changed product direction: ≥ 1
- Investor conversations enabled by traction data: 3+
- Confidence level (self-reported): 3/10 → 7/10
| Attribute | Detail |
|---|---|
| Age | 25-34 |
| Background | Full-time engineer at startup or big tech |
| Location | Anywhere — has a stable day job |
| Revenue | $0-2K MRR (side income) |
| Tech Stack | Whatever they're comfortable with |
| GitHub Bio | "Building things on weekends." |
Jordan has a day job they don't hate but don't love. They dream of financial independence through side projects. They've started 10+ side projects but finished 2. Life gets in the way — work deadlines, social commitments, Netflix.
They need a lightweight system that fits into 5-10 hours/week, provides accountability, and makes the most of limited time. They don't need complex project management. They need a shipping ritual that keeps them moving.
- Ship consistently on weekends — 1 meaningful commit every week
- Finish projects — Get to "launched" not just "started"
- Learn in public — Build an audience as a byproduct
- Eventually quit day job — Reach $5K MRR from side projects
| Pain | Current Behavior | Emotional State |
|---|---|---|
| Inconsistent progress | Ships 3 weeks in a row, then nothing for 2 months | Guilty, cyclical |
| No accountability | Nobody knows if they shipped or not | Isolated, easy to quit |
| Forgotten context | Comes back after 2 weeks, forgets where they were | Confused, demotivated |
| Scope creep | "I'll just add this one feature..." | Overwhelmed, never ships |
| Weekend decision fatigue | Sits down Saturday, doesn't know what to work on | Wastes half the day |
Sunday Morning: loopkit loop (The Ritual)
- 9 AM, coffee in hand
- Runs
loopkit loop— 2-minute data aggregation - Sees: "Week 3 summary: 2 tasks done, 1 open, shipping score 67%"
- AI recommendation: "Ship the MVP this week — it's 80% done"
- Accepts → writes reason for override (if any) → gets BIP post draft
- Posts to Twitter: "Week 3 of building [X]. This week: shipping the MVP. No more tweaks."
- Feels accountable — public commitment made
Saturday: loopkit track
- Opens laptop, runs
loopkit track - Sees exactly what needs to happen: 3 tasks, 1 is snoozed
- Works for 4 hours, commits
[#2] Finalize auth flow - Task auto-closes. Shipping score now 80%.
Sunday Afternoon: loopkit ship
- MVP is live on Vercel
- Runs
loopkit ship→ "What did you ship?" → "MVP with auth + core feature" - AI generates launch copy
- Uses HN draft, posts to Show HN
- Ship log saved — milestone captured
Week 4: loopkit pulse
- Gets first real user feedback via shared form
--add "User said onboarding is confusing"- Only 3 responses — raw mode shows them directly
- Feels validated — real people are using it
- Consecutive weeks with ≥ 1 commit: 8+ (2-month streak)
- Projects taken from init to shipped: 2+/year
- Hours per week on side project: 5-10 (sustainable)
- Side project revenue growth: +20% month-over-month
| Dimension | Sarah (Solo Founder) | Marcus (Indie Hacker) | Alex (First-Timer) | Jordan (Side Project) |
|---|---|---|---|---|
| Primary Goal | Product-market fit | Shipping velocity | Validation | Consistency |
| Time Available | 40-60 hrs/week | 40-80 hrs/week | 20-40 hrs/week | 5-10 hrs/week |
| Risk Tolerance | Low (can't afford another failure) | High (fails fast, moves on) | Low (career transition) | Medium (safety net from day job) |
| Tech Skills | Expert | Expert | Beginner-Intermediate | Intermediate |
| Key Command | init (rigor) |
track (speed) |
init (clarity) |
loop (accountability) |
| Pain Intensity | High (stuck) | Medium (scattered) | High (lost) | Medium (inconsistent) |
| Pay willingness | High (needs this to work) | Medium (sees value) | High (saves consulting fees) | Low-Medium (nice to have) |
- Manages 10-person teams with Jira, Confluence, and quarterly OKRs
- Needs resource allocation, Gantt charts, and stakeholder reports
- Why not LoopKit: Too lightweight, not collaborative, no enterprise features
- Manages 5+ client projects simultaneously
- Needs client visibility, billing integration, and team assignment
- Why not LoopKit: Single-founder focused, no multi-user support (yet)
- Has $2M in the bank, 5-person team, board meetings
- Needs OKR tracking, investor updates, and hiring pipelines
- Why not LoopKit: Too early-stage, not team-oriented enough
- Codes for fun, no intention of monetizing
- Wants to learn new frameworks, build portfolio pieces
- Why not LoopKit: The validation and shipping pressure would feel like work, not play
For Product Decisions: Before adding any feature, ask: "Which persona needs this most, and what pain does it solve?"
For Copywriting: When writing CLI output, UI text, or marketing copy, write for Sarah (our primary persona). She's smart but overwhelmed. She needs clarity, not cleverness.
For Prioritization: Features that help multiple personas ship more consistently should be prioritized over features that only help one persona in edge cases.
For Onboarding: The first-run experience should assume Alex (first-timer) level of familiarity. Power users like Marcus will figure out advanced features on their own.
Last updated: April 2026