> For the complete documentation index, see [llms.txt](https://docs.lyslabs.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.lyslabs.ai/the-platform.md).

# The Platform

## Layered design philosophy

Everything in LYS is modular. Each component is purpose-built for a specific part of the data pipeline, and they all connect seamlessly.

| **Layer**        | **What It Does**                                 |
| ---------------- | ------------------------------------------------ |
| Raw Data Capture | Collect Solana blocks, transactions, logs        |
| Decoders         | Parse and classify events from protocol activity |
| Aggregators      | Compute trends, metrics, signals                 |
| APIs & WSS       | Deliver insights in real time or on demand       |
| Apps / Agents    | Built on the data: Solexys, dashboards, alerts   |

These layers can be consumed individually or as a whole - from bare metal streaming to AI-ready insight generation.

## Core components

### **1. Indexer**

* Built on Solana Geyser + custom fallbacks
* Streams blocks every 400ms
* Fully stateless, event-driven design

### **2. Decoders**

* Protocol-specific logic: Pump.fun, Raydium, Meteora, LaunchLab, and others
* Extract swap actions, bonding curves, governance events, wallet flows etc.
* Normalized into shared schemas for consistency

<img src="https://paper.dropboxstatic.com/static/img/ace/emoji/1f517.png?version=8.0.0" alt="link" data-size="line">[ See Decoder Structure →](/decoders.md)

### **3. Aggregators**

* Real-time and historical metrics computed on decoded events
* Designed for:
  * Price signals (ATH, volume spike)
  * Wallet stats (PnL, buy/sell ratio)
  * Token diagnostics (creator behavior, rug flags)

<img src="https://paper.dropboxstatic.com/static/img/ace/emoji/1f517.png?version=8.0.0" alt="link" data-size="line"> [Explore Aggregations →](/contextualized-data.md)

### **4. Data Distribution methods**

* WebSocket Streams
* REST APIs for structured querying (coming soon)
* GraphQL (coming soon)

<img src="https://paper.dropboxstatic.com/static/img/ace/emoji/1f517.png?version=8.0.0" alt="link" data-size="line"> [Use the APIs →](/apis-and-streams.md)

### **5. Applications**

* Dev Portal: Portal for managing API keys, subscription and API documentation along querying live/historical Solana data
* Fully programmatic access also available for power users

{% hint style="info" %}
Apps are clients of the same APIs exposed to you.
{% endhint %}

## Platform flow, from block-to-bot

Let’s walk through a simplified real-time journey:

{% stepper %}
{% step %}

### Indexer

Captures block #38,510,203 from Solana mainnet
{% endstep %}

{% step %}

### Decoders detect events (examples)

* LaunchLab token creation
* Pump.fun liquidity init
* 4 buys, 1 sell, and a bonding curve jump
  {% endstep %}

{% step %}

### Aggregators compute

* Volume spike > 2x rolling avg
* ATH hit
* 3 unique buyers, 1 is tagged “smart wallet”
  {% endstep %}

{% step %}

### Streams / APIs

* Bots subscribed to /tokens/alerts WSS stream are notified in <20ms
* Solexys agent receives JSON bundle for token assessment
* Dashboard updates with new chart data
  {% endstep %}
  {% endstepper %}

## Integration options

| **You Are...**         | **You Should Use...**                    |
| ---------------------- | ---------------------------------------- |
| Building a trading bot | Subscribe to real-time alerts via WSS    |
| Doing quant research   | Query historical APIs with full filters  |
| Training an AI model   | Use structured JSON from decoders        |
| Creating dashboards    | Embed REST calls or use Sandbox directly |
| Monitoring token risks | Filter by creator, sniper, bundle flags  |

## Composability by design

Every layer of the LYS platform is:

* **Modular:** Swap in your own models, frontends, or indicators
* **Low-latency:** Engineered for millisecond decision-making
* **Schema-first:** All decoded data adheres to a common structure
* **Graph-ready:** Enables multi-hop reasoning and clustering
* **AI-optimized:** Built to power both supervised models and agents. Data analytics and interpretation AI powered

{% hint style="info" %}
Use what you need. Ignore what you don’t. The platform adapts to you.
{% endhint %}

## What’s Next

Now that you understand how the platform is structured, it’s time to dig into the technical architecture - the services, pipelines, and infrastructure that make this all work.

<img src="https://paper.dropboxstatic.com/static/img/ace/emoji/1f449.png?version=8.0.0" alt="backhand index pointing right" data-size="line"> Continue to Architecture