Home • About • Projects • Blog • Resources • Contact
Update: TinyLaunch is now Launchling! Same idea, new name and better experience.
🛠️ Building TinyLaunch
A lightweight AI tool to help non-technical founders take the first step — and a playground for me to learn how to build inclusive, agentic products.
Problem
Getting started is often the hardest part of building something new — especially for non-technical or first-time founders.
There’s a flood of startup advice out there, but it’s often overwhelming, overly generic, or assumes a level of confidence, capital, and clarity that many people just don’t have (yet).
What’s more, most tools and resources tend to speak to a particular audience: confident, tech-savvy, often male. I wanted to build something more inclusive - a tool that encourages anyone with an idea (however scrappy or uncertain) to take a small, focused first step.
The hypothesis: with the right tone, guidance, and AI scaffolding, more people could see themselves as builders and have the confidence to get started.
Learning Goals
While I’d love TinyLaunch to evolve into something genuinely useful, it’s also a learning project.
Through building it, I’m aiming to:
- Explore what agentic, user-centred AI tooling can look like
- Practice scoping and shaping early-stage experiments using LLMs
- Improve my skills in prompt design, orchestration, and UX writing
- Get more hands-on with AI tools and infrastructure
- Learn how to build lean, safe, no-code AI experiences without over-engineering
Solution
TinyLaunch is a lightweight, no-sign-up tool that generates a tailored startup plan based on a user’s confidence, experience, goals, and constraints.
The goal isn’t to create a business plan or spit out a polished pitch deck — it’s to nudge people over the starting line with something practical, doable, and encouraging.
So far, it provides:
- A one-page, AI-generated starter plan
- Clear steps to validate the idea and get early signals
- Tool recommendations matched to budget/confidence
- Optional follow-up and simple tracking (early stage)
Tech / Tools used
So far I’ve used:
- OpenAI API for generating personalised plans
- Tally for onboarding form and user input collection
- Framer for visual prototypes
- Notion for hosting supplementary resources
Next steps will be to use the following
Key insight
This has been a great project to get started on because in a way I’m dogfooding it by building it.
As I started mapping risks, assumptions, and success criteria, I realised these needed to be built into the product itself. This early reflection loop helped shape a stronger MVP and reminded me how valuable it is to build with your user in mind — especially when that user is you.
Project Updates
8 Jul 25
Status - Work in progress!
Right now, it’s a functioning alpha with:
- Guardrails to limit GPT-4 token usage
- Early feedback capture
- Privacy and GDPR compliance baked in
Next steps:
- Finish end-to-end workflow
- Improve plan UX and visual design
- Add versioning and “I started building” tracking
- Explore idea clustering and success metrics
Learnings so far
- A lot can be done with a lightweight AI stack and no backend
- Picking a ‘meta’-project that could help me learn through dogfooding was an excellent move
Also: it’s been a lot of fun!
🪂 Try it now
Turn your idea into a tiny, tangible prototype — with a little help from Launchling.
👉 Try Launchling – no signup, no jargon, just a gentle push forward.