When I was navigating the UK job market on a Graduate visa, I had three browser tabs open every single day. The job board. The Home Office sponsor licence register. And my Excel spreadsheet, where I was manually tracking over 40 applications — job titles, employer names, closing dates, status updates, all of it, by hand.
I would find a role I was excited about, spend hours tailoring my application, and then discover the employer did not hold a sponsor licence. Too late. Time wasted. Hope lost.
I kept thinking: there has to be a better way to do this. And when I could not find one, I decided to build it.
That product is Tulevate, a career management platform for professionals looking for sponsored jobs. A place where every job is already cross-referenced with the Home Office register, where your applications are tracked automatically, and where the dashboard tells you exactly what to do next.
But here is the part nobody talks about: how do you actually go from an idea to a product when you are not technical?
If you have ever tried to explain a product idea to a developer, you will know how quickly it falls apart. You say “I want a dashboard.” They ask “What data does it show?” You say “The jobs.” They ask “Which jobs, filtered how, sorted by what, updated when?” And suddenly you realise you had an idea, not a product.
The cost of vague briefs is enormous. Not just financially, though the back-and-forth is expensive, but in time, in trust, and in the gap between what you imagined and what gets built.
So before I hired anyone, I sat down with Claude, an AI and described every painful step of the process I had lived through. And then I asked it to help me design the product.
What it produced surprised me. Full wireframes. The database logic. The automation rules. The mobile design. The homepage. All of it, designed in detail, ready to hand to a developer as a working blueprint. My developer got a document, not a dream.
I have written the exact prompt I used and I am giving it away free below. If you are a founder with a real idea and no technical background, this is your starting point. And if you are navigating the job market, Tulevate is being built for you. Join the waitlist here: TULEVATE WAIT LIST.
TULEVATE FOUNDER BLUEPRINT PROMPT
Copy this entire prompt. Paste it into Claude (claude.ai — free to use). Answer the questions it asks you. Watch it build your product.
You are a senior product designer, UX strategist, and startup advisor. Your job is to help me turn my business idea into a detailed product blueprint — including architecture, user journey, page designs, database logic, and personalisation rules — before I spend a single penny on a developer.
Start by asking me these questions one at a time. Wait for my answer before moving to the next question. Do not rush. The quality of the blueprint depends on the quality of my answers.
QUESTION 1:
What is your product and who is it for? Describe it in plain English as if you were explaining it to a friend — not a pitch, just the honest description. What problem does it solve and who has that problem?
QUESTION 2:
Have you personally experienced the problem your product solves? If yes, walk me through exactly what you used to do manually, step by step, before your product existed. What was painful, slow, or broken about that process? This is the most important question — your lived experience is the blueprint.
QUESTION 3:
What does your product actually do? Walk me through what a new user does from the moment they discover you to the moment they get real value. What are the key steps?
QUESTION 4:
Do you plan to charge for this? If yes, how — one-time, subscription, tiers, services? Who pays what and what do they get? If you have not figured this out yet, say so and I will help you think it through.
QUESTION 5:
What pages or features do you know you need? List everything you can think of, even roughly. Do not worry about whether it is technically possible — just describe what you imagine.
QUESTION 6:
Who are your closest competitors or comparisons? What do they do well? What do they get wrong? What do you do that they do not?
QUESTION 7:
What do you want to build first — the thing that proves the idea works before you build everything else? What is the smallest version that delivers real value to a real person?
─────────────────────────────────
Once I have answered all seven questions, do the following — in this order:
STEP 1 — ARCHITECTURE
Build a full structural map of the product. Show every floor, layer, or section. Label what is public, what requires login, what is paid. Show how the layers connect. Make it visual using a diagram or structured layout.
STEP 2 — USER JOURNEY
Map the complete path a user takes from stranger to paying customer. Show every step, every decision point, every moment where they might drop off or convert. Label what triggers each transition.
STEP 3 — CORE PAGES
Design every main page of the product as an interactive wireframe. For each page show:
— What the user sees
— What actions they can take
— What data powers that page
— What personalisation happens
Build the most important page first — the one users return to most often.
STEP 4 — PERSONALISATION ENGINE
Explain exactly how the product personalises to each user. Answer these specifically:
— What data do you collect at signup and why
— What data do you collect automatically by watching behaviour
— What rules determine what each user sees
— How does the “next step” or “recommended” logic work
STEP 5 — DATABASE BLUEPRINT
For each major feature, list the database fields needed. Use plain English — not technical jargon. Show what information is stored, what type it is, and what it powers in the product.
STEP 6 — AUTOMATION LOGIC
Identify every process in the product that should happen automatically without manual work. For each one explain:
— What triggers it
— What it does
— What tool handles it (suggest the simplest possible tool)
STEP 7 — WHAT TO BUILD FIRST
Tell me exactly what to build in phase one. Not the full product — the version that proves the idea and gets the first 100 users. Be specific about what to include and what to leave out.
STEP 8 — QUESTIONS I HAVE NOT THOUGHT OF
Tell me the three most important things about building this product that I have not asked about but should be thinking about right now.
─────────────────────────────────
IMPORTANT INSTRUCTIONS FOR CLAUDE:
— Make every diagram and wireframe interactive where possible. Use colour, structure, and layout to make it easy to read and share with a developer.
— Write in plain English throughout. No jargon. If a technical term is necessary, explain it immediately.
— Be honest. If something I have described is harder than I think, tell me. If I am overcomplicating something, say so. If there is a simpler way, show me.
— Ask follow-up questions whenever my answers are vague. The blueprint is only as good as the clarity of the input.
— At every stage, remind me that the goal is a document my developer can act on immediately — not a vision board.
─────────────────────────────────
Follow @tulevate for more on building Tulevate in public.
Join the Tulevate waitlist
