{% extends "base.html" %}

{% block title %}IAToolkit - Foundation{% endblock %}

{% block styles %}
    <link rel="stylesheet" href="{{ url_for('static', filename='styles/landing_page.css') }}">
    <link rel="stylesheet" href="{{ url_for('static', filename='styles/documents.css') }}">
{% endblock %}

{% block content %}
<div class="container-fluid p-0">
    {% include '_static_header.html' %}
</div>

<div class="article-container">
    <header class="article-header">
        <h1 class="article-title">IAToolkit Mini-Project: Scope, Phases, and Timeline</h1>
        <p class="article-subtitle">
            A three-month, high-learning project to deploy a real AI assistant integrated with your corporate data.
        </p>
    </header>

    <div class="article-content">

    <h2>1. Overview</h2>

    <p>
      The <strong>IAToolkit Mini-Project</strong> is a structured, three-month initiative designed to deploy
      an internal AI assistant powered by IAToolkit and connected to real company data, documents, and workflows.
      It provides a practical, low-risk way for organizations to explore AI with tangible results and measurable progress.
    </p>

    <p>
      Instead of aiming for a massive rollout, the mini-project focuses on a single business unit or client,
      ensuring that the scope remains manageable while delivering a production-grade assistant and deep organizational learning.
    </p>

    <h2>2. Goals of the Mini-Project</h2>

    <ul>
      <li>Deploy a functional, company-specific AI assistant based on IAToolkit.</li>
      <li>Enable natural-language access to SQL databases and structured business information.</li>
      <li>Integrate internal documents using retrieval-augmented generation (RAG).</li>
      <li>Implement custom Python tools that allow the assistant to take real actions.</li>
      <li>Create a clean multi-tenant foundation for future assistants.</li>
      <li>Train internal teams on prompt design, schema modeling, and domain context.</li>
    </ul>
<h2>3. Timeline and Phases (≈ 3 Months)</h2>

<p>
  A complete mini-project spans <strong>three months</strong>, divided into four structured phases.
</p>
<div class="iat-stair-timeline">

  <div class="iat-stair-step step-1">
    <div class="iat-stair-header">
      <span class="iat-stair-phase-badge">Phase 1</span>
    </div>
    <h4 class="iat-stair-title">Setup &amp; First Company</h4>
  </div>

  <div class="iat-stair-step step-2">
    <div class="iat-stair-header">
      <span class="iat-stair-phase-badge">Phase 2</span>
    </div>
    <h4 class="iat-stair-title">Model Data &amp; Knowledge</h4>
  </div>

  <div class="iat-stair-step step-3">
    <div class="iat-stair-header">
      <span class="iat-stair-phase-badge">Phase 3</span>
    </div>
    <h4 class="iat-stair-title">Tools &amp; Workflows</h4>
  </div>

  <div class="iat-stair-step step-4">
    <div class="iat-stair-header">
      <span class="iat-stair-phase-badge">Phase 4</span>
    </div>
    <h4 class="iat-stair-title">Pilot &amp; Production</h4>
  </div>

</div>

<div class="iat-intelligence-bar">
  <div class="iat-intelligence-labels">
    <span>Intelligence available</span>
    <span class="iat-intelligence-end">More AI value</span>
  </div>
  <div class="iat-intelligence-track">
    <div class="iat-intelligence-fill"></div>
  </div>
  <div class="iat-intelligence-marks">
    <span>Simple queries</span>
    <span>Ad-hoc prompts</span>
    <span>Custom Python tools</span>
    <span>RAG &amp; domain knowledge</span>
</div>
</div>

<h3>Phase 1 – Setup, Exploration, and Company Creation (≈ 3 weeks)</h3>

<p>
  This phase establishes the foundation of the project by installing the toolkit, exploring its structure,
  aligning the scope, and creating the initial Company module.
</p>

<ul>
  <li>Install and run IAToolkit locally.</li>
  <li>Explore the <code>sample_company</code> and understand the architecture: prompts, schemas, context, tools.</li>
  <li>Define project scope and target questions with stakeholders.</li>
  <li>Create the new Company directory (e.g. <code>companies/acme_corp/</code>).</li>
  <li>Configure <code>company.yaml</code> with LLM, embeddings, tools, and branding.</li>
  <li>Connect SQL databases and validate the first generated queries.</li>
  <li>Set up basic authentication and internal testing access.</li>
</ul>

<h3>Phase 2 – Modeling Knowledge and Data (≈ 3–4 weeks)</h3>

<p>
  This phase focuses on translating business knowledge into structured schemas and documents.
</p>

<ul>
  <li>Write YAML schema descriptions of the key tables and relationships.</li>
  <li>Add business rules and domain explanations in <code>context/</code> as Markdown.</li>
  <li>Test LLM-generated SQL and iterate for accuracy.</li>
  <li>Refine terminology and structure with domain experts.</li>
  <li>Ensure reliable access to the datasets used in the first use cases.</li>
</ul>

<h3>Phase 3 – Tools, Workflows, and Documents (≈ 3–4 weeks)</h3>

<p>
  Once the assistant understands your data, the next step is enabling it to act and reason.
</p>

<ul>
    <li>Create reusable prompts for recurring business tasks.</li>
  <li>Implement Python tools (reports, notifications, calculations, API integrations).</li>
  <li>Define tool schemas in <code>company.yaml</code> for structured function calling.</li>
  <li>Load internal documents into the vector store for RAG.</li>
  <li>Validate multi-step workflows combining SQL, RAG, and tools.</li>
</ul>

<h3>Phase 4 – Pilot and Production Rollout (≈ 2–3 weeks)</h3>

<p>
  The last phase validates the assistant with real users and prepares it for production.
</p>

<ul>
  <li>Deploy a pilot version to a controlled user group.</li>
  <li>Monitor usage using IAToolkit’s history and feedback tools.</li>
  <li>Refine prompts, schemas, and tool behavior based on real-world interactions.</li>
  <li>Finalize production deployment and internal documentation.</li>
</ul>

<p>
  By the end of the three-month cycle, the organization has both a working AI assistant and the
  internal knowledge to continue expanding it to new datasets, workflows, and business units.
</p>
    <h2>4. Expected Outcomes</h2>

    <ul>
      <li>A fully functional AI assistant integrated with your real business data.</li>
      <li>A documented set of schema files, prompts, tools, and context documents.</li>
      <li>A maintainable Company module ready for expansion.</li>
      <li>Hands-on experience with prompt engineering, SQL modeling, and RAG.</li>
      <li>A clear roadmap for scaling AI to additional business units.</li>
    </ul>

    <h2>5. Team and Technical Requirements</h2>

    <h3>Team</h3>

    <ul>
      <li><strong>1 Python Developer</strong> (responsible for configuration and tools)</li>
      <li><strong>1 Domain Expert</strong> (validates answers, defines rules, provides examples)</li>
      <li>Optional: <strong>DevOps engineer</strong> for cloud setup and monitoring</li>
    </ul>

    <h3>Technical Stack</h3>

    <ul>
      <li>Python 3.12+</li>
      <li>PostgreSQL for IAToolkit’s internal database</li>
      <li>Redis for session handling</li>
      <li>Local or S3-compatible object storage for documents</li>
      <li>A Git repository for your Company code and configuration</li>
      <li>A cloud environment capable of hosting a Flask application</li>
    </ul>

    <h2>6. Why This Approach Works</h2>

    <p>
      The three-month structure provides the right balance: long enough to deploy something meaningful,
      but short enough to stay focused and manageable. It ensures controlled experimentation,
      real-world validation, and internal learning before scaling to larger initiatives.
    </p>

    <blockquote>
      “The goal is not only to deliver an assistant, but to help the organization understand how to
      design and operate AI systems grounded in their own data.”
    </blockquote>

    <h2>7. Next Steps</h2>

    <p>
      Because <strong>IAToolkit is fully open source</strong>, organizations can begin immediately:
    </p>

    <ul>
      <li>Clone the repository and experiment locally.</li>
      <li>Use this mini-project structure as an implementation guide.</li>
      <li>Adapt schemas, prompts, and tools to your own data and workflows.</li>
    </ul>

    <p>
      If you're exploring AI for internal use, the IAToolkit Mini-Project offers a clear, practical,
      and structured way to get started.
    </p>

    {% include '_static_footer.html' %}

   </div>
</div>
{% endblock %}