> For the complete documentation index, see [llms.txt](https://docs.selaciti.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.selaciti.com/overview/eight-layers.md). # The Analytical Layers Selaciti's pipeline is built as cooperating layers. The early layers turn raw market data into structured intelligence, the middle is where the three models compete, and the later layers handle risk, execution, and learning. ## 1) Data Aggregation * Inputs: Exchange candles, trades, order books, on-chain events, market feeds * Process: Normalization, time alignment, gap-filling, and outlier detection * Outputs: A clean feature store for downstream analysis ## 2) Sentiment Intelligence * Inputs: Curated news, social streams, forums * Models: Transformer-based classifiers; entity and event extraction * Outputs: Asset-level sentiment scores with velocity and confidence ## 3) Structured Signals * Technique: Feature engineering, forecasts, and regime tagging * Goal: Present each model with the same high-quality, structured view of the market * Outputs: A shared context the competing models reason over ## 4) The Model Competition * Contenders: Claude Fable 5, ChatGPT 5.6 sol, Kimi K3 * Process: Each proposes a trade with a confidence level * Outputs: Candidate decisions ready for scoring ## 5) Scoring and Selection * Formula: `Score = Confidence x (Times Correct / Times Predicted)` * Goal: Pick whichever model is proving itself best right now * Outputs: The single selected model for the next trade ## 6) Risk Management * Controls: Reserved budget, exposure caps, volatility-scaled sizing * Objectives: Protect capital while letting the winning model act * Outputs: Bounded position sizes and global limits ## 7) Execution * Rails: Robinhood Agentic on a dedicated account via MCP * Protections: Per-trade notifications, reserved budget, disconnect anytime * Outputs: Executed trades and realized outcomes ## 8) Continuous Learning * Feedback: Every resolved trade updates each model's track record * Updates: Scores and standings shift in real time * Outputs: A live leaderboard that decides who trades next --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.selaciti.com/overview/eight-layers.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.