Eight Buildix Features Most Traders Never Open

Most traders open the screener, sort by a column, look at three pairs and close the tab. That is maybe fifteen percent of what is actually there, and the parts that go unopened are the ones designed to work while nobody is watching a screen.

Here are eight, in rough order of how much time they save.

Saved Filters Turn the Screener Into a Monitor

A screener answers the question you ask right now. A saved filter answers it continuously.

Build a condition set in the screener, funding above a threshold, VPIN above a level, open interest change over a window, then save it with a name. The filter keeps evaluating against all 530 pairs and can notify by Telegram, email or webhook when a pair enters the set. Each filter carries its own cooldown in minutes so a pair oscillating around your threshold does not produce forty messages.

Conditions combine with AND or OR logic, which is the part people miss. A filter for high funding OR extreme VPIN gives you noise. High funding AND rising OI AND price above the 4 hour EMA200 gives you three or four names a week that are actually worth looking at.

Alerts Can Be Composite and One Shot

The alert builder is separate from saved filters and works per symbol. Beyond simple price levels it supports composite mode, where several conditions must be true together, plus a timeframe, a direction filter and a one shot toggle.

One shot matters more than it sounds. An alert that fires once and disarms itself is the correct configuration for a level you only care about the first time it breaks. Leaving alerts armed and repeating is how people end up muting the bot, which defeats the entire setup.

Alerts can also be bound to a saved filter, so the screener condition and the per pair alert become the same object rather than two things you maintain separately.

Wallet Follows Fire on Open, Not After the Move

Whale tracking that reports what already happened is a newsletter. Following a specific wallet address and getting notified when it opens a position is a signal.

Add an address, give it a label so you remember why you added it, and choose whether to be notified on position open, on close, or both, and through which channel. The tracker holds the last known positions for each followed wallet, so the notification is a state change rather than a periodic snapshot.

The practical use: after a whale case study or a large flow shows up in the screener, follow that address. Attribution on Hyperliquid is public, so the same wallet is traceable across every subsequent position it takes.

The Trading Journal Stores the Signal You Entered On

The journal records symbol, direction, entry and exit price, size, PnL in dollars and percent, notes and tags. That is standard. The part that is not standard is that it also stores the signal score and signal direction at the moment of entry.

That single field is what turns a journal from a diary into a dataset. After forty trades you can answer whether your entries above a confidence of 60 outperformed the ones below it, and whether the trades you took against the signal direction were the disasters you suspect they were. Almost nobody keeps this data, which is why almost nobody knows their own edge.

Watchlists Sync Across Devices

Named watchlists hold a symbol set and live on the account, not in browser storage. Open the same watchlist on a phone and it is identical to the desktop.

The same applies to chart preferences. Indicator layout, timeframe defaults and layer choices persist per account, so a chart set up on a laptop opens the same way on mobile rather than resetting to defaults.

API Keys Are Available Below the Top Tier

Every account tier above free can generate API keys with a label, and each key carries its own daily request budget that resets on a rolling window. The dashboard shows requests used today and last used timestamp per key, which is how you find the script that is quietly burning the quota.

Label keys by what they do rather than by date. A key called python-backtest and a key called telegram-bot can be revoked independently when one of them misbehaves.

The Changelog Is a Real Page

Every shipped feature, fix and data change is logged with a date, a tag and a version. It is the fastest way to find out whether something changed underneath you, which matters when a strategy that worked last month behaves differently this month.

The AI Strategy Advisor Runs on Your Own Key

Bring a key from OpenAI, Anthropic, Google, Groq, Mistral or a local Ollama instance and the advisor reads live orderflow context into whichever model you chose. Your key, your provider, your data path, and no markup on inference.

The setup that pays for itself is the least glamorous one: a saved filter with three conditions, Telegram delivery, a 60 minute cooldown, and a journal entry logged on every trade it produces. Build that once at buildix.trade/screener and it keeps running whether or not you open the tab.