Draconic: AI That Actually Reads Markets

You can’t process everything. Five timeframes, momentum, options flow, microstructure. All while deciding if your setup is real or a trap.

You know what happens when you miss something.

Prop desks have teams for this. Analysts watching flows, quants running models, someone always tracking what you’re not. You’re doing it alone. Draconic is that team. It reads the full picture, validates your thinking, and flags what you’re not seeing. In real-time.

Our thesis

We don’t believe AI will ever fully automate trading. Markets are adversarial, adaptive, and full of edge cases that break any fixed system.

But AI can do something humans can’t: synthesize dozens of signals across multiple timeframes in real-time, without fatigue, without bias, without missing things because you were focused elsewhere. That’s what we built.

Draconic is currently powered by frontier models. The difference is what we’ve built around it. Deep context engineering, trading principles, pattern recognition frameworks, and structured market data so the AI interprets it the way an experienced trader would. You still make the decisions.

What it actually does

Ask Draconic about a setup, and it doesn’t just fetch data. It tells you what’s supporting the trade AND what could invalidate it. Multi-timeframe alignment, volume confirmation, hidden divergences, nearby liquidity zones.

A few examples:

Morning prep: “Market status? Indices, momentum, key levels across 5m/15m”. You get a read on whether the day favors longs, shorts, or sitting out.

Before entry: “Planning to go long AAPL at 180. Acceptance or rejection zone? Any hidden risk?” Draconic flags what you might be missing: exhaustion signals, divergences, or distribution patterns that aren’t obvious on the chart.

Position management: “I’m long BTC from 67,500, SL at 66,800. What should I watch?” Real-time monitoring for momentum shifts or level tests that might change your thesis.

Opportunity scanning: Instead of flipping through 30 charts, ask “Top 3 momentum setups right now with entry and risk levels.”

Where it works

Full coverage (near real-time, refreshed continuously):

  • Indian indices and F&O: NIFTY, BANKNIFTY, SENSEX, including options data, PCR, gamma walls, max pain
  • Top 200 NSE/BSE stocks: RELIANCE, TCS, HDFCBANK, INFY, etc.
  • Global commodities: Gold, Silver, Crude etc.
  • US stocks: Top 100 including AAPL, NVDA, TSLA, MSFT, GOOGL
  • Crypto: BTC, ETH, major alts
  • Forex: EURUSD, GBPUSD, USDJPY and other major pairs
  • Global indices: SPX, NASDAQ, DAX, NIKKEI

Limitations worth knowing:

  • Small-cap Indian stocks (outside top 200) have limited data
  • Options analytics only available for top 50 stocks and indices
  • US stocks beyond top 100 have basic coverage

Try it

What you’re testing today is the foundation. We’re building toward a fully agentic trading terminal: analysis, execution, and an agent that remembers your setups, learns your edge, and personalizes how you actually trade.

Early Access is open. Full launch in January.

Draconic specializes in providing decision-support primarily for global intraday traders and short-term swing traders, NOT long-term investors.

Sign up here.

Website - https://draconic.ai/

Use code: TQNA100

We’re posting here because we want feedback from traders who’ll actually push the product. Tell us what’s useful. Tell us what’s missing. Tell us where it’s wrong.

2 Likes

Too heavy UI and/or slow website.

Cool :sunglasses:

How does it compare to the FOSS AI-Trader, include proofs please. :100:

https://ai4trade.ai/

3 Likes

Hey thanks for checking it out. The UI is a conscious design choice, and of course it will keep improving over time. In terms of website being slow could you tell us specifically what the issue was?

I think it took more memory than tradingview and there was noticeable lag in a old small system.

Thanks for sharing this. Doesn’t look like the product is targeting the same use case as us. Seems like they are comparing different LLMs performance on trading decisions. We are creating an AI analyst that has customized tools, logic which will augment your ability to make trading decisions. It doesn’t take trading decisions on its own.

Got it. The tool itself (on hunt.draconic.ai) however should use much lesser memory than TradingView

This product is completely unreliable for stock evaluation. It does not provide real insight and simply agrees with whatever sentiment you express. If you say a stock is doing well, it finds random reasons to justify that. If you say it is doing badly, it shifts its stance to match. There is no independent analysis and no data-driven reasoning. The numbers it gives are often incorrect, and it cannot even generate a proper 5-day analysis. Everything feels like generic LLM output with no real understanding of market behavior. Overall, it is a worthless tool for anyone trying to assess whether a stock is actually performing well or poorly.

1 Like

Any subscription based or hows the pay structure features currently

Looks like you have created a new tradingqna account just to share your feedback about Draconic - we feel special :slight_smile:
Why don’t you enlighten us with the real product screenshots of your query + the response from Draconic - would love to see the specifics. Also happy to see those same queries with any other product out there to compare and learn.

It’s free to use for now