Before agentic AI, going from an idea to MVP to PROD used to take a lot of time.
At least for me, it usually meant sketching UI ideas on paper, starting to code, getting an MVP ready, then building a website, and finally deploying. Depending on the idea, that could take anywhere from 1 to 3 to even 6 months.
Recently I found the Tiingo API for stock and dividend data and realized its free tier was more than enough for my personal needs. I had already been paying $9.99/month for a dividend tracker app, so I thought… why not just build my own?
This time I used agentic AI—mainly Claude CLI, and sometimes CODEX.
I went from idea to MVP in a day. By day 2, I had posted about it on Reddit for beta testers and had 47 people helping test it. By day 4, I had a working website and was already spending more time on making the app PROD ready instead of spending weeks trying to get version 1 off the ground.
The app is called Finance Forest: Portfolio.
We are living in some crazy times 🙂
So how do I do it?
1. Create a dummy project and open Claude CLI or CODEX in terminal
I start with a clean project and a very simple statement of the idea. In this case: build an iOS app that uses the Tiingo API to track holdings, dividend income, and future payments.
2. Create a Documentation folder right away
Before any serious coding starts, I create a Documentation folder. This is where I keep product notes, MVP scope, feature ideas, API decisions, and implementation phases. For Finance Forest: Portfolio, that included things like multi-portfolio support, dividend calendar, 12-month forecast, and iCloud sync.
3. Start in plan mode — most important step — never start by coding
This is the biggest lesson for me. Never begin with “build the app.”
Start with “help me think through the app.”
For example, I asked Claude to help me figure out what the actual MVP should be for a dividend-focused portfolio app. Not everything needed to be in version 1. I wanted clarity on what was essential now versus what could wait.
4. Discuss the idea properly and have Claude store the plan in the Documentation folder
This is where the real back-and-forth happens. For Finance Forest: Portfolio, the discussion was around questions like:
What should the onboarding look like if the user needs a Tiingo API key?
Should the app focus on real-time prices? No, end-of-day was enough.
What are the core views? Portfolio summary, dividend calendar, forecast, and charts.
What makes this useful enough for someone to switch? Seeing dividend income clearly and simply.
Once the discussion was good, I had Claude save the plan into docs.
5. Turn that plan into an implementation checklist with phases
After that, I asked Claude to convert the idea into build phases.
Something like:
Phase 1 – app structure and models
Phase 2 – Tiingo API integration
Phase 3 – portfolio and holdings screens
Phase 4 – dividend calendar and forecast
Phase 5 – charts, import/export, and polish
That immediately makes the project feel real and manageable.
6. Start coding phase by phase
This part matters. I do not ask for everything at once.
For example, I would first focus only on fetching stock prices and dividend history from Tiingo. Then move to showing holdings. Then build the dividend calendar. Then the 12-month forecast. Breaking it up gives much better output and makes reviews much easier.
7. Test and commit after every phase
After each phase, I test and commit. If Tiingo integration works, commit it. If the dividend calendar works, commit it. If the forecast logic looks right, commit it. This keeps progress clean and avoids one giant messy code dump.
This keeps progress clean and avoids one giant messy code dump. It also gives me safe checkpoints along the way. If something gets messed up in the next phase, I can always go back to the previous working version instead of trying to untangle everything at once. That alone makes experimenting much less stressful.
8. Once the app is working, use Claude to build the website
After the MVP was usable, I used the same process to build a website for it. The site needed to explain what Finance Forest: Portfolio does, why it is useful for dividend investors, how Tiingo setup works, and how people could join testing.
What makes this part much easier is that Claude already has the context it needs. It has seen the plan, the feature list, the positioning, the onboarding flow, and the overall purpose of the app from the documentation created earlier. So instead of starting from scratch again, I can ask it to turn what already exists into a website. That makes the process much faster and usually gives a better first version too.
9. Use Claude for all the “words” work
This part is underrated. Once the product starts taking shape, I use Claude for App Store description, subtitle ideas, onboarding copy, and even Reddit launch posts.
For Finance Forest: Portfolio, I used Claude for much more than coding. It helped me think through the idea, build the MVP phase by phase, create the website, write the App Store copy, and even get the app in front of beta testers.
That is the part that still blows my mind a bit.
Before, just getting to that stage used to take months. Now I can go from idea to something real and testable in days, learn from actual users, and then decide what deserves the deeper PROD effort.
Crazy times 🙂