How to Build a Blockify Dataset from Sales Collateral (Docs to Blocks)

Become the sales representative whose answers are always right and always ready. Blockify transforms your messy sales decks and fragmented documents into a single source of truth, empowering you to close deals faster and with unwavering confidence.

This comprehensive guide is designed for sales operations, sales enablement teams, and product marketing professionals. Its goal is to provide a detailed, step-by-step walkthrough for creating a curated, trusted Retrieval-Augmented Generation (RAG) dataset using Iternal Technologies' patented Blockify technology. You'll learn how to ingest sales decks and various documents, deduplicate content, write critical questions, approve trusted answers, and effectively tag metadata. This process positions your Artificial Intelligence (AI) with an astounding 78 times greater accuracy compared to vanilla RAG, ensuring governance by design, and includes a robust review workflow, clear block naming conventions, and a streamlined update cadence.

1. Understanding the "Why": The Problem with Traditional Data for Artificial Intelligence

Before we dive into the solution, let's understand the challenges that Iternal Technologies’ AirgapAI and Blockify technology are designed to solve. When people talk about Artificial Intelligence, especially Large Language Models (LLMs) like chat applications, they often think about powerful tools that can answer almost any question. However, when these tools try to answer questions using your specific company data, they often run into significant problems:

• Artificial Intelligence Hallucinations: Imagine asking a system about your company's latest product specifications, and it confidently gives you incorrect information or makes up details that aren't true. This is an Artificial Intelligence hallucination. It happens because LLMs, especially when trained on general internet data, lack the precise, verified knowledge of your internal documents. If you can't trust the Artificial Intelligence once, you can't ever truly trust it, leading to lost productivity and credibility.
• The Challenge of "Bring Your Own Data" (BYOD): Many cloud-based Artificial Intelligence solutions allow you to upload your own documents. However, these documents are often messy—full of redundancies, outdated information, and inconsistencies. Standard Large Language Models struggle to sift through this noise, leading to inaccurate responses, even with your own data.
• Cost Implications of Inaccurate Artificial Intelligence: Every time an Artificial Intelligence hallucinates or provides an incorrect answer, your team wastes time validating the information, correcting errors, or re-running queries. This erodes trust, slows down workflows, and ultimately costs your organization money in lost productivity and missed opportunities.
• Data Sovereignty and Security Concerns with Cloud Artificial Intelligence: Perhaps the biggest hurdle for organizations adopting Artificial Intelligence is data security. Sending your proprietary sales collateral, financial data, or sensitive customer information to a third-party cloud service, even a "trusted" one, introduces significant risks. Regulatory compliance (like Health Insurance Portability and Accountability Act (HIPAA) or General Data Protection Regulation (GDPR)) and internal security policies often prevent the use of cloud-based Artificial Intelligence solutions for confidential information, leading to a standstill in Artificial Intelligence adoption.

These challenges highlight a critical need: an Artificial Intelligence solution that is not only powerful and easy to use but also inherently secure, accurate, and cost-effective, especially when working with your most valuable, sensitive data.

2. Introducing AirgapAI and Blockify: Your Solution for Trusted Artificial Intelligence

Iternal Technologies developed AirgapAI and its core Blockify technology to directly address the pitfalls of traditional Artificial Intelligence deployment, providing a groundbreaking approach to secure, accurate, and cost-effective Artificial Intelligence.

2.1. What is AirgapAI? A Local, Secure Large Language Model Platform

AirgapAI is a revolutionary, locally installed, on-device Large Language Model (LLM) platform that brings the power of conversational Artificial Intelligence directly to your Personal Computer (PC). Think of it as your own personal, highly secure chat application, but with a critical difference:

• 100% Local, No Cloud Required: AirgapAI runs entirely on your client device, such as a Dell Artificial Intelligence Personal Computer, completely eliminating the need for external network connections to the cloud for its core operations. This means your data never leaves your device, providing unparalleled privacy and security—an essential advantage for all organizations, particularly those with stringent security requirements like government agencies, defense contractors, and financial institutions.
• Ultimate Privacy and Security: Because AirgapAI operates in an "air-gapped" environment, there's no network traffic "in" or "out" of your device. This design ensures absolute data sovereignty, protecting your sensitive information from external breaches and keeping it within your existing corporate security perimeter.
• Cost-Effective and Ownership: Unlike cloud alternatives that burden you with ongoing subscription fees and unpredictable token charges, AirgapAI is sold as a one-time perpetual license per device. This means you own your Artificial Intelligence, avoiding the recurring costs associated with solutions like Microsoft CoPilot or ChatGPT Enterprise, often at a tenth (or even less) of the cost.
• Runs on Your Artificial Intelligence Personal Computer: AirgapAI is optimized to leverage the full power of modern Artificial Intelligence Personal Computers, utilizing the Central Processing Unit (CPU), Graphics Processing Unit (GPU), and Neural Processing Unit (NPU) for maximum performance and efficiency. It integrates seamlessly into standard imaging workflows, making deployment across your fleet as straightforward as any other application.
• Key Features for Business Users:
  • Standard Chat Interface: Interact with the Artificial Intelligence just like popular cloud-based chat applications.
  • Guided Workflows: Access Quick Start workflows tailored for specific roles (e.g., procurement, legal, marketing) with pre-configured prompts and relevant datasets.
  • Entourage Mode: A unique feature allowing you to interact with multiple Artificial Intelligence personas simultaneously, providing diverse perspectives from different datasets for complex decision-making or content creation.

2.2. What is Blockify? The Engine of Artificial Intelligence Accuracy

While AirgapAI provides the secure, local platform for Artificial Intelligence, Blockify is the patented technology that ensures the Artificial Intelligence running on it is exceptionally accurate. Blockify is the ultimate data management solution for Large Language Models at scale, designed to transform your chaotic corporate data into a pristine, reliable source of truth.

• How it Solves the Hallucination Problem: Blockify systematically processes your data to eliminate the messiness that causes Artificial Intelligence hallucinations. Instead of feeding a Large Language Model raw, unorganized documents, Blockify distills information into highly precise, verified data "blocks."
• The "Docs to Blocks" Concept: This powerful concept refers to Blockify's ability to ingest vast amounts of unstructured data (your "docs") and intelligently condense and structure it into concise, modular "blocks" of information. This process ensures that every piece of data the Artificial Intelligence accesses is accurate, relevant, and up-to-date.
• Achieve 78 Times (7,800%) Improvement in Accuracy: By leveraging Blockify, you can reduce the original data size by as much as 97.5% and, remarkably, improve the accuracy of your Large Language Models by 7,800%. This is a game-changer for building trust and achieving reliable outcomes with your Artificial Intelligence.

3. Getting Started with AirgapAI: Installation and First Launch (High-Level)

Installing AirgapAI is designed to be straightforward, akin to installing any standard desktop application. While this article focuses on the Blockify workflow, here's a high-level overview of the initial setup:

3.1. System Requirements and Prerequisites

AirgapAI Chat is optimized for modern Personal Computers, especially those with Artificial Intelligence capabilities.

• Central Processing Unit (CPU): Minimum 8 cores; Recommended 8 Cores/16 Threads or better.
• Random Access Memory (RAM): Minimum 16 Gigabytes; Recommended 32 Gigabytes or more.
• Disk Space: Minimum 10 Gigabytes free (Solid State Drive); Recommended 50 Gigabytes Non-Volatile Memory Express (NVMe).
• Graphics Processing Unit (GPU) (Integrated or Dedicated): Minimum 4 Gigabytes Video Random Access Memory (VRAM) (2024 or newer); Recommended 8 Gigabytes VRAM or more.
• Operating System (OS): Microsoft Windows 11 with the latest patches.
• Permissions: You will need appropriate security permissions to install software on your device.

3.2. Downloading and Installing the Application

1. Obtain the Installer: Your Information Technology (IT) department will provide the latest AirgapAI installer package, typically a ZIP archive.
2. Extract the Files: Right-click the ZIP file and select "Extract All..." to a writable folder.
3. Run the Installer: Double-click the AirgapAI Chat Setup.exe file.
4. Follow the Wizard: Accept the license agreement, choose to create a desktop shortcut, and click "Install." If prompted by your Operating System's security, choose to "Allow" or "Run anyway."

3.3. First-Launch Onboarding Wizard

Upon launching AirgapAI Chat for the first time, an onboarding wizard will guide you through essential configurations:

1. Profile and Chat Style: Enter a display name and select your preferred chat interface style.
2. Uploading the Core Large Language Model: You will be prompted to upload a core Large Language Model. Browse to the /models/ folder within your extracted installer. Choose a model suited to your hardware (e.g., Llama-1B for integrated Graphics Processing Units or low-power devices, Llama-3B for more powerful integrated Graphics Processing Units or dedicated Graphics Processing Units).
3. Uploading an Embeddings Model: Next, upload an embeddings model, typically Jina-Embeddings.zip from the /models/ folder. Embeddings models are crucial for Retrieval-Augmented Generation, as they help the Artificial Intelligence understand the context and meaning of your data.
4. Adding Sample or Custom Datasets: While not strictly necessary for initial setup, you can add sample datasets or your Blockify-processed custom datasets here. We will focus on creating these custom datasets in the next section.
5. Finish Onboarding: Once all models are uploaded, click "Continue" to launch AirgapAI Chat.

3.4. Initial Model Benchmarking

After the initial setup, AirgapAI Chat will offer to benchmark your hardware. It's highly recommended to "Run Benchmark" (takes approximately two minutes) to measure your device's tokens per second and inference speed. This ensures optimal performance and allows you to expand the Artificial Intelligence's context window for longer, more detailed conversations.

4. The Core Workflow: Building a Blockify Dataset from Sales Collateral (Step-by-Step)

This section details the critical process of transforming your sales collateral into a highly accurate, trusted Retrieval-Augmented Generation (RAG) dataset using Blockify. This is where you achieve the 78 times improvement in Artificial Intelligence accuracy.

4.1. Goal: Curated, Trusted Retrieval-Augmented Generation Dataset

Your objective here is to create a refined, single source of truth from your company's sales and marketing documents. This dataset will empower AirgapAI to deliver precise, trusted answers to specific questions, directly impacting sales effectiveness and customer engagement.

4.2. Step 1: Data Ingestion – Gathering Your Sales Collateral

The first step is to identify and organize the raw materials for your Artificial Intelligence's knowledge base.

• Identify Relevant Documents: Collect all pertinent sales and marketing collateral. This typically includes:
  • Sales presentations and decks
  • Requests for Proposals (RFPs) and their responses
  • Product specification sheets
  • Frequently Asked Questions (FAQs) documents
  • Customer testimonials and success stories
  • Legal disclaimers or compliance documents
  • Internal training materials for sales teams
• Supported File Formats: Blockify is highly versatile and natively ingests a wide array of document types:
  • Text files (.txt)
  • HyperText Markup Language (.html)
  • Portable Document Format (.pdf)
  • Microsoft Word documents (.docx)
  • Microsoft PowerPoint presentations (.pptx)
  • Graphic files (images, from which text can be extracted or described).
  • For video content, Blockify can extract still frames or transcribe audio as needed.
• Recommendation for Best Results: For optimal performance and to take full advantage of Blockify's hierarchical metadata and taxonomy framework, we recommend that your customer data be curated into relevant categories. For example, group documents by specific product lines, business units, or customer segments. This pre-organization helps Blockify understand the contextual relationships between different pieces of information.

4.3. Step 2: The Blockify Process – From "Docs" to "Blocks"

This is the core of Blockify's magic, transforming disparate documents into a structured, highly accurate dataset.

• Creating a New Task within Blockify:
  • Open the Blockify application.
  • You will initiate a new task, for example, naming it "AirgapAI Sales Enablement" or "Q3 Product Launch Materials."
• Uploading the Document:
  • Within the new task, you will upload your selected sales documents. You can upload them individually or in batches.
  • Blockify will begin its ingestion and processing sequence.
• Automatic Extraction of Key Blocks:
  • As Blockify processes your documents, it intelligently extracts and condenses key information into modular "blocks" of data.
  • Each block is meticulously structured to ensure clarity and precision:
    • Name (Displayed in Blue): This is a concise title or topic that quickly identifies the content of the block. For instance, "AirgapAI Cost Benefits" or "Blockify Accuracy Metrics."
    • Critical Question (Bold and Italicized): This is the key query that a customer, or your internal team, might ask. It's designed to directly target the essential information within the block. For example, "What is the cost advantage of AirgapAI compared to Microsoft CoPilot?"
    • Trusted Answer (Light Gray Text): This represents the distilled, accurate, and approved response to the critical question. This answer is meticulously crafted to avoid the pitfalls of outdated or redundant data, ensuring it is always a single source of truth.
• Emphasize Data Reduction: This process is incredibly efficient. Blockify can reduce the original data size of your documents by as much as 97.5% (down to just 2.5% of the original content), making it incredibly efficient for Large Language Models to process.
• Metadata Tagging for Security and Governance: Each block is automatically tagged with rich metadata. This includes:
  • Classification: Designating the sensitivity or type of information (e.g., Public, Internal, Confidential, Classified).
  • Permissions: Assigning user access rights to specific blocks, crucial for zero-trust environments and ensuring only authorized personnel can access certain data.
  • Classification Levels: Further defining security tiers or departmental relevance. This metadata ensures that your Artificial Intelligence adheres to your organization's security policies and data governance frameworks, allowing you to gate access to sensitive datasets by individual user role or persona.

4.4. Step 3: Human Review and Curation Workflow

While Blockify is highly intelligent, the "human in the loop" remains crucial for ultimate data governance and trust.

• Why Human Review is Crucial: After Blockify's initial ingestion and block creation, these blocks are sent for a quick human review. This step is vital for:
  • Validation: Ensuring the automatically extracted critical questions and trusted answers accurately reflect the intended message.
  • Updating Outdated Content: Flagging and updating any content (e.g., a "2019 pricing model") before it impacts Artificial Intelligence responses, preventing the dissemination of stale or incorrect information.
  • Refining Messaging: Optimizing the trusted answers for clarity, conciseness, and brand voice.
• Updating or Approving Messaging: Designated subject matter experts (e.g., product managers, legal counsel, sales leadership) can review, edit, or approve each block, ensuring the dataset is a fully vetted, accurate, and official source of information.

4.5. Step 4: Dataset Management and Updates

Your sales collateral is dynamic, and so should be your Artificial Intelligence's knowledge base.

• How Updated Datasets are Pushed to Local Devices: As new documents are Blockified or existing blocks are updated, these revised datasets can be pushed to the local devices running AirgapAI. This is seamlessly managed via standard Information Technology image management applications such as Microsoft Intune, Ivanti Endpoint Manager, or similar tools, just like you would update any other file or application.
• Continuous Improvement of the Dataset: This process enables continuous improvement. You can regularly update your Blockify datasets to reflect the latest product changes, market trends, or sales strategies, ensuring your AirgapAI is always working with the most current and accurate information available.

5. Leveraging Your Blockify Dataset within AirgapAI Chat

Once your Blockify dataset is created, reviewed, and deployed, you can immediately harness its power within the AirgapAI Chat application. This is where your sales team becomes the rep whose answers are always right and always ready.

5.1. Retrieval-Augmented Question and Answer (QA) with Blockify Datasets

This is the primary way your team will interact with your curated data, ensuring highly accurate responses.

1. Switch to the AirgapAI Application: Launch AirgapAI Chat from your desktop shortcut.
2. Toggle Your Dataset On: In the sidebar of the AirgapAI interface, you will see a list of available datasets. Select your newly created Blockify dataset (e.g., "Iternal Technologies Enterprise Portfolio Overview"). A toggle switch will allow you to activate it, making it the primary source for the Artificial Intelligence's knowledge.
3. Ask a Specific Question: Now, you can ask questions directly related to the information contained within your sales collateral.
   • For example, with your "Iternal Technologies Enterprise Portfolio Overview" dataset selected, you might prompt: "What is Iternal Technologies?" or "What is AirgapAI?"
   • Or, in a sales context: "What are the key benefits of the AirgapAI solution for data privacy?"
4. How the Retrieval-Augmented Generation Engine Works:
   • When you ask a question, AirgapAI's RAG engine first queries your selected Blockify dataset.
   • It intelligently identifies and ranks the top data blocks that are most relevant to your question.
   • The Large Language Model then synthesizes a coherent, trusted answer based only on the information from those verified blocks.
   • Crucially, the system will also show citations back to the specific blocks it used to formulate the answer, allowing for quick validation and transparency.

5.2. Role-Based Workflows

AirgapAI enhances productivity by providing Quick Start workflows that are pre-configured for different roles and tasks.

• Accessing Workflows: These workflows are easily accessible on the Workflow Bar, typically located below the new chat window.
• Tailored Prompts and Curated Datasets:
  • For instance, a "Sales Proposal - Cover Letter" workflow could be selected.
  • When activated, this workflow might automatically select specific sales-focused datasets and provide a pre-engineered prompt like "Write a cover letter for a new Dell Artificial Intelligence Personal Computer proposal."
  • Users can then upload supporting documents or simply click "Generate" to receive a fully crafted output, which can then be copied to their clipboard.
• Personalized Experiences: Because the AirgapAI application is tied to the user's profile on login, you can have multiple users on the same device, each leveraging the application with their own isolated experiences and role-specific datasets. This is configured per user profile through your standard Information Technology image and provisioning process.

5.3. Entourage Mode (Multi-Persona Chat)

Entourage Mode is a truly unique feature that allows for advanced ideation and multi-perspective analysis, especially useful for complex scenarios.

• What it Is: Entourage Mode allows users to interact with multiple Artificial Intelligence personas simultaneously. Each persona can be configured with expertise from different Blockify datasets, providing distinct viewpoints on a single query.
• How it Works:
  1. Select a Workflow: Choose an Entourage Mode Quick Start workflow from the new chat page.
  2. Configure Personas: In "Advanced Settings → Personas," you can define and configure various Artificial Intelligence personas (e.g., Marketing Specialist, Legal Advisor, Technical Expert, Financial Analyst).
  3. Example Use Case (Business): When preparing a complex proposal, you could have a "Marketing Persona" (tuned from marketing collateral), a "Legal Persona" (from legal contracts), and a "Technical Support Persona" (from product specifications) all weigh in on the same question. Each persona will lend different perspectives from their respective Blockify datasets, providing a comprehensive, multi-faceted response.
  4. Example Use Case (Defense/Intelligence): In a defense or intelligence scenario, you could configure one persona as a Central Intelligence Agency (CIA) analyst and another as a military tactician. The CIA analyst persona is set up with expertise in intelligence gathering, target package details, and sensitive data interpretation. The military tactician persona is tuned to provide insights on ground operations, combat strategies, and tactical decision-making. Users can simultaneously ask the same question and receive distinct answers from each expert, giving a multi-perspective view on complex issues and supporting high-stakes decision-making.
• Recommended Prompt for Demonstrating Entourage Mode: "I am launching a new product called AirgapAI. It is a 100% local chat Large Language Model solution that is 1/10th the cost of other solutions with more capabilities. What do you think? Please answer in short sentences."

6. Advanced Capabilities and Ongoing Management

AirgapAI is built for flexibility, scalability, and ease of management within an enterprise environment.

6.1. Model and Dataset Management

• Bring Your Own Model (BYOM): AirgapAI supports a "Bring Your Own Model" approach, meaning you can use any of the popular, common open-source Large Language Models available (e.g., Llama, Mistral, DeepSeek) or even bring your own fine-tuned custom Large Language Models. If a needed model isn't pre-quantized (optimized for local deployment), Iternal Technologies' engineering team can package and deploy it as a service. This ensures the flexibility to adapt to evolving Artificial Intelligence needs.
• Context-Window Expansion: After completing the initial model benchmark, you can expand the Artificial Intelligence's context window (the amount of information it can "remember" and process in a single conversation) up to 32,000 tokens by going to "Settings → Model Settings" and adjusting the slider.

6.2. Ongoing Updates and Maintenance

• Streamlined Update Cadence: AirgapAI's update cadence is designed to synchronize with your typical Operating System or enterprise software update cycle. Whether pushing application updates, security patches, or new Blockify datasets, your Information Technology (IT) department can deploy new versions through familiar image management solutions (e.g., Microsoft Intune).
• Update Manager: AirgapAI includes a built-in Update Manager. You can configure it to receive updates from a local server (for enhanced security and control) or a cloud source in "Settings → Updates." Your Information Technology team can even modify the update file server location directly within the updaterConfig.json file.

6.3. Training and Support

Iternal Technologies is committed to ensuring your team's success with AirgapAI.

• Comprehensive Training: We offer a 30-minute introductory demonstration, followed by personalized training sessions as an add-on service. This ensures your team quickly becomes proficient with the application.
• Online Enablement Page: Our dedicated online enablement page serves as a rich resource, featuring step-by-step video tutorials, extensive Frequently Asked Questions (FAQs), detailed user guides, and troubleshooting tips.
• Dedicated Support: Our customer success team is readily available for follow-up calls, additional workshops after initial deployment, and to answer any questions via support@iternal.ai.

7. Conclusion: Empower Your Workforce with Trusted, Secure, and Cost-Effective Artificial Intelligence

AirgapAI, powered by Blockify technology, offers a transformative solution for organizations seeking to leverage Artificial Intelligence without compromising security, accuracy, or budget. By following this detailed guide, you can confidently build and deploy highly effective Retrieval-Augmented Generation datasets from your sales collateral, empowering your sales operations, enablement, and product marketing teams.

Here’s a recap of the unparalleled value AirgapAI brings:

• A Fast Artificial Intelligence Win: Experience immediate business value with an easy-to-deploy Artificial Intelligence solution that integrates seamlessly into your existing workflows.
• Robust Cost Savings: Own your Artificial Intelligence with a perpetual license, dramatically reducing costs compared to expensive, recurring cloud subscriptions—often at 1/10th or even 1/15th of the price.
• Unparalleled Data Security: Operate in a 100% local, air-gapped environment, ensuring your sensitive data never leaves your device and adheres to the strictest data sovereignty and privacy requirements.
• Virtually Eliminate Artificial Intelligence Hallucinations: Our patented Blockify technology refines your data inputs into a pristine, single source of truth, leading to an astonishing 78 times (7,800%) improvement in Large Language Model accuracy. This builds unwavering trust in your Artificial Intelligence.

AirgapAI provides a trusted, secure, and cost-effective Artificial Intelligence experience designed for the Artificial Intelligence Personal Computer, enabling your workforce to unlock unprecedented productivity and make informed decisions with confidence.

Ready to transform your sales enablement and empower your team with a truly secure and accurate Artificial Intelligence assistant?

Download the free trial of AirgapAI today at: https://iternal.ai/airgapai