EV Battery Health Report

EV Battery Health Report X Lovable

Bootstrapped project built and shipped using Lovable, GPT, figma in a less than a sprint

Problem statement: EV buyers lack an independent, data informed ways to assess real world battery health and range risk before purchase, leading to uncertainty and poor decision making. The current solutions require booking a car retailer or garage service and paying for a technician to run a diagnostic. My tool addresses the inconvenience by offering a remote service that gives an estimated report to speed up the process and help customers with buying and selling decisions. A report that downloads in seconds.

Product scope summary

I identified the trust gap in the used EV market and designed a focused, single model MVP that translates mileage and age into a clear, statistically bounded battery degradation estimate. I defined the product scope, built the end to end user flow, implemented strict payment gating via Stripe, and created legally conservative safeguards to protect both users and the business. By prioritising precision over breadth, I delivered a credible, monetisable tool that balances data transparency, risk management, and premium UX execution.

Vibecoding workflow

Figma: Defined the core concept, mapped user flows, and built a design system canvas in parallel.

OpenAIGPT: Validated the product idea, stress tested modelling logic, drafted legal safeguards, and formalised system rules.

Lovable: Built and shipped the live MVP, implemented payment gating, and deployed the production ready experience.

Stripe: Integrated secure checkout and payment confirmation logic.

Merchant bank: Enabled transaction settlement and financial processing.

GoogleStitch: Rapidly explored layout structures for phase 2 in the fastest way and explored an upgraded design system to scale the product

Discord: Collected early feedback, validated assumptions, and iterated quickly post launch.

Learnings

Overall I solved a problem, ideated, designed, built and shipped a new product in record speed time. agentic AI platforms designed for full stack software development are serious productivity hacks and immense fun.Here are some themes I spent time problem solving and learning to refine my methods on, in a shorter time boxed period

Prompt maturity: learning to write all inclusive rules that served front and back end

Lovable will not execute a high standard of UI polish – this must remain in the hands of craft to refine, polish and judge aesthetics

Inconsistencies will occur throughout the flows unless I later created rule and logic patterns for holistic replacements and updates

Stripe integration had delays in its backend/ security. It is not as ‘fast and swift’ as Lovable portal experience might suggest

Technical knowledge will help you unblock, debug and ship faster