Crypto sector research workspace with a personal watchlist

Build me a crypto research workspace that our analysts open to browse the market by sector rather than one coin at a time. Market data comes from CoinGecko and every analyst's personal watchlist is stored in Airtable.

The landing view is a sector heatmap built from the CoinGecko Coin Categories operation. Render one tile per category showing the category name, its total market cap, its 24 hour change, and the names of its top constituent coins. Size or color the tiles by market cap and change so a glance tells you which sectors are green and which are red today. Sort the strongest and weakest sectors to the top rather than listing categories alphabetically.

Clicking a sector drills into that sector's coins. Use the Coins Markets operation filtered by that category id, ranked by market cap, and show a table with coin name and ticker, current price, 24 hour and 7 day change, market cap, and 24 hour volume. Page through results rather than loading everything at once. Every row has an add to watchlist action.

Across the top of the landing view, show a trending strip built from the Trending operation, listing the coins people are searching for most in the last 24 hours. Each trending coin is clickable and opens that coin the same way a sector row does, so a spike in search interest is one click away from being watchlisted.

Analysts should never need to know exact coin ids. Adding a coin goes through a search box backed by the CoinGecko Search operation: the analyst types a name or a ticker, the app shows matching coins with their icon, name, and symbol, and picking one resolves the underlying coin id behind the scenes.

The watchlist lives in an Airtable table with one row per coin per analyst. The columns are the coin id, coin name, ticker, the sector it came from, the analyst who added it, a thesis note, a conviction rating, and the date added. Adding a coin calls Create Records and stamps the row with the signed in analyst's identity. Removing a coin calls Delete Records. Loading the watchlist calls List Records with a filter on the analyst field so each person sees only their own rows, never the whole team's. This per analyst filtering matters: the point is that each analyst keeps their own list rather than everyone sharing one.

Each watchlist row is editable inline. The analyst can write a short thesis note explaining why they are watching the coin and set a conviction rating, and saving calls the Airtable Update Record operation against that row. Keep the note short form, a couple of sentences rather than a document, and show when it was last updated so stale theses are obvious.

There is also a compare view. The analyst selects up to four coins from their watchlist and the app puts them side by side. For each coin, draw a 90 day price series from the Market Chart operation, and pull market cap, trading volume, circulating supply, and total supply from the Get Coin by ID operation. Normalize the price lines so coins at very different price points can be compared on shape rather than absolute value, and show the thesis note and conviction rating alongside each column so the analyst compares their own reasoning next to the numbers.

Practical notes. Show all money values in US dollars by default but make the display currency easy to change. CoinGecko's free tier is rate limited to roughly 30 calls per minute, so fetch per view and cache category and market responses briefly instead of refetching on every render. Airtable allows about five requests per second per base, so batch watchlist writes where you can. Handle the empty states properly: a new analyst with no watchlist rows should see a prompt to search for their first coin, and the compare view should explain that it needs at least two coins selected.