Raj Parikh.

Available

PlainStock PlainStock

Understand your stocks, not just track them.

Live data, AI generated analysis, and peer comparisons turned into plain language reports you can read in seconds.

Role

Solo designer and builder. Product, interface, backend, database, API integrations, and the prompt engineering behind every report.

Launch PlainStock
The PlainStock dashboard: a sortable table of held stocks with live price movement across three timeframes The AI Brief for Berkshire Hathaway open in front of the dashboard, showing similar stocks, related articles, and a 'Was this report helpful?' prompt
Outcome

Designed and built solo, with AI as my engineering team. The hard part wasn't getting AI to write a report. It was deciding what it was allowed to say.

Part 01: The problem

More consumption than comprehension

I use Fidelity to manage my portfolio, but getting a clear picture of one stock takes multiple tabs, pages, and clicks. Every session felt like more consumption than comprehension.

Try it live

It opens a sandbox: you can add up to 3 stocks, and nothing you change touches the real list.

Role
Solo designer and builder
Design and build
Claude, Antigravity, Google Stitch, Figma
APIs and infrastructure
Gemini API, Finnhub, Polygon, Supabase, Vercel
Status
Used by friends and family through private invites

Part 02: The product

Live data, AI reports, and peer analysis in a single view

PlainStock offers a complete read on your portfolio without leaving the page. Each stock shows real time price movement across three timeframes, with one click access to an AI generated report and peer comparison.

The main view

Dashboard

The main view pulls live market indices and stock performance into one sortable table. Stocks can be added, removed, or moved to a future prospects watchlist.

The dashboard: market indices across the top, then a sortable table of held stocks with price and movement across three timeframes
The main table, sortable across three timeframes.
The report

AI Brief and Analyzer

Each AI Brief turns a stock's position, recent movement, and analyst signals into a plain language report. The Analyzer suggests similar stocks to consider and explains why.

The AI Brief for Eli Lilly: overview, market position with figures, and an analyst take with a buy, hold and sell bar
One stock, written out in plain language.
The Stock Analyzer for Intel: suggested peers AMD and Nvidia, one already in Future Prospects, and a plain language explanation of why
Peers to consider, with the reasoning.

Part 03: The approach

A system designed before it was built

I mapped the data sources, connections, and costs before building, so each build decision traced back to a product decision.

  • Decision

    A table built for scanning

    Comparing stocks across timeframes is the core task, so the main view is one sortable table.

  • Decision

    Words, not charts

    Reading a chart takes financial know-how. A plain sentence doesn't.

  • Decision

    Holdings apart from prospects

    Watching what you own and weighing what you might buy are different mindsets, so they're different lists.

How I worked with AI

AI did the building. Claude and Antigravity wrote the backend, database, and integrations, work I couldn't have done alone.

I did the deciding. What the product is for, what a good report looks like, and every rule about what the AI can and can't do.

01

Planning

  • Identify gaps
  • Core features
02

System Thinking

  • API mapping
  • Data sources
  • Caching and security
03

UI Exploration

  • Figma
  • Prototyping with AI
04

Report Algorithm

  • Prompt engineering
  • Gemini integration
  • Plain language output
05

Eval + Rubric

  • Define quality standards
  • Tested across 6 stocks
  • Iterated on failures
06

Output

  • Live demo
  • Rule checks on every report
  • Reader feedback
01

Planning

  • Identify gaps
  • Core features
02

System Thinking

  • API mapping
  • Data sources
  • Caching and security
03

UI Exploration

  • Figma
  • Prototyping with AI
04

Report Algorithm

  • Prompt engineering
  • Gemini integration
  • Plain language output
05

Eval + Rubric

  • Define quality standards
  • Tested across 6 stocks
  • Iterated on failures
06

Output

  • Live demo
  • Rule checks on every report
  • Reader feedback

Part 04: Trust

Trust was the hardest thing to build

Early reports rambled, cited junk sources, dropped sections, and read differently on every visit. It took about 13 versions and four decisions to get one I trusted.

Decision 1

Define “good” before tuning anything

Every report had to meet four rules:

  • Every fact links to a real source. Blogs, content farms, and other AIs are blocked.
  • Plain but not childish: written for a smart 14-year-old with no finance background.
  • The same sections every time. None ever goes missing.
  • It never invents a number.

I scored every report against these rules while building it, and fixed each gap.

The owner's report quality page: each report with its rule checklist result and reader votes. Older versions of MU and WDC fail with specific reasons, including non-US stocks and a blocked Substack source, while current reports pass.
My quality page. Older report versions fail these rules. Current ones pass.
  • Decision 2

    AI proposes, code verifies

    The AI writes. Code supplies the facts and sources, and checks every suggestion against real market data.

  • Decision 3

    Predictable and safe to run

    One fixed structure, shared across users, refreshed only when the news changes, with hard spending caps.

  • Decision 4

    Checked after launch

    Every live report runs the rule check and gets reader votes. I review the flagged ones.

A tool I built because I wanted it to exist, and kept working on until I trusted what it told me.

All work Next case study Bank of America