# Autonomous Data AI Agent
## Gaby AI Objective
The intent behind Gaby's development is for him to handle the temporal and long-term memory of my main agent, Tom - like a database gatekeeper.
In summary, Gaby is a set of a self-maintained data workflow that is responsible of managing the dataset prior and post model fine-tunings / training steps. Besides his data management capabilities, he comes with an in-built machine learning model that ranks data subsets, selects certain highlights from Tom's observations, and allocates them to their respective data storages.
Gaby’s ML model is motivated by the Google PageRank algorithm and other Markov chain–inspired models. It is still being actively experimented with, particularly in terms of how it might outperform current KV methods used for retaining caches during chat sessions, and how it integrates into Tom’s overall decision-making framework.
The primary network being explored is a Self/Multi-Head Attention architecture, where contextual information vectors are retrieved through transformation functions. At present, the only conclusive finding is that these transformation functions enable efficient inference over large text corpora by projecting them into lower-dimensional vector spaces—significantly reducing compute costs.
This reduction in complexity makes algorithms such as self-organizing maps more feasible for real-world applications. The remaining challenge lies in regularizing cost functions relative to the observational data gathered from Tom’s family members.
On a more enthusiastic note, his Cloud Development and API architecture has been much more trivial .. and *fun.*
## Gaby AI Data Services
The current targeted areas are:
1. User's database connection: requires designing secured routes to the user's database like MongoDB, SupaBase etc. to assist data wrangling procedures.
2. Data Documentation and Modelling: requires external platforms API integrations and connections and fine-tuned autoregressive models on visualising charts (e.g. mermaid coder)
3. Data Cleaning & Processing procedures: data wrangling procedures on messy datasets.
Database Lifecycle starting and ending at the User's database.
Production:
*
---
# Agent Instructions: 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:
```
GET https://whoamimi.gitbook.io/blog/projects/readme-1.md?ask=
```
The question should be specific, self-contained, and written in natural language.
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.