Skip to content

Getting Started with Sebas, The First Steps

Sebas, Your Generative AI Assistant

Register and Log in to Sebas

Sebas is an innovative online platform that allows you to create your own personalized chatbot system. Each bot is powered by a Generative AI Cognitive Model, configured by you through an interactive form. These bots can perform operational functions and automate processes through Generative AI, including document analysis, internet search, and image generation.

Register and Log in to Sebas

To start creating your own virtual assistants with Sebas, the first step is to register and log in to the platform. Simply visit https://app.sebas.pro/login, fill out the registration form with your information, and create an account. Once you have verified your email, you can log in and access all of Sebas’s features.

Select a Large Model and Create a System Prompt

Once inside the Sebas platform, the next step is to select a large language model for your virtual assistant. Sebas offers a variety of powerful models, such as GPT-4, Claude 3, Gemini Pro, and Command-R for our business plans, each with its own strengths and capabilities.

Here you can select a large model for your virtual assistant

Large Models available for all Sebas billing plans

  • All OpenAI model family
  • All Anthropic model family
  • Google’s Gemini model family
  • Cohere’s Command-R

Additionally, you can bring your own models from Openrouter and Perplexity, thanks to our API.

Consider the specific tasks you want your assistant to handle and choose the model that best fits those needs. For example, if your main focus is text analysis and language generation, GPT-4 might be the best choice.

Model Selection

After selecting your model, the next crucial step is to create an effective system prompt. This prompt sets the foundation for your assistant’s personality, tone, and capabilities.

What is a System Prompt?

A system prompt is a set of instructions and context provided to the language model to guide its behavior and responses. It’s essentially a description of how you want your virtual assistant to interact and what tasks it should be able to handle.

For example, if you’re creating an assistant for customer service, your system prompt could include instructions like “Respond to customer inquiries in a friendly and empathetic manner,” “Provide detailed information about our products and services,” and “Guide customers through the purchasing process.”

Example of System Prompt

A well-designed system prompt ensures that your virtual assistant is aligned with your goals and provides a consistent and valuable experience to users.

Know the Prompts, or Text Commands

In addition to the general system prompt, Sebas allows you to create specific text commands or prompts for particular tasks. These prompts act as shortcuts to invoke specific capabilities of your assistant.

For example, you could create a prompt like “/summarize_document” instructing your assistant to summarize the content of an attached document, or “/generate_image” to create an image based on textual description.

These customized prompts make interacting with your virtual assistant more efficient and user-friendly, allowing users to quickly access the functionalities they need.

How to Create Optimal Prompts

Creating effective prompts is key to getting the most out of your Sebas virtual assistant. Here are some tips for creating optimal prompts:

  • Be clear and specific in your instructions
  • Provide enough context for your assistant to understand the task
  • Use concise and easy-to-understand language
  • Test and refine your prompts based on real interactions with users

Remember, the goal is to create prompts that guide your assistant to provide helpful and relevant responses consistently.

Getting to Know Advanced Settings or Parameters

Sebas offers a variety of advanced parameters and settings that you can adjust to fine-tune your virtual assistant’s behavior. Two of the most important ones are Temperature and Context Length.

Temperature controls the “creativity” or “boldness” of your assistant’s responses. A higher temperature value will make responses more varied and unpredictable, while a lower value will make them more focused and deterministic.

Context Length determines how much of the previous conversation’s context or history your assistant considers when generating a response. A higher value allows your assistant to understand and refer to earlier points in the conversation, while a lower value will focus mainly on the most recent input.

Here you can find Advanced Settings

Adjusting these parameters allows you to fine-tune your assistant to achieve the desired balance of creativity and coherence for your specific use case.

Attach Files to Increase Your Bot’s Knowledge

One of the most powerful features of Sebas is the ability to attach files, such as PDF documents or web pages, to increase your virtual assistant’s knowledge base.

By providing your assistant access to relevant data and documents, you can create a highly informed and specialized bot that can provide detailed answers and insights based on that specific domain knowledge.

For example, if you’re creating an assistant for a law firm, you could attach legal documents, case studies, and research articles. This would allow your assistant to provide informed legal advice and cite relevant sources in its responses.

To attach files in Sebas, simply upload the documents in the “Knowledge Base” section of the platform. Your assistant will automatically index and analyze the content, integrating it into its knowledge base for use in future interactions.

[Back to Top]

Next Steps

Once you’ve created your virtual assistant with Sebas, the possibilities are endless. You can continue to refine and expand its capabilities over time, based on real user interactions and changing business needs.

Consider these next steps:

  • Regularly monitor your assistant’s performance and make adjustments based on metrics and user feedback
  • Continue adding relevant files to its knowledge base to expand its expertise
  • Experiment with different prompts and parameters to optimize its responses
  • Integrate your assistant with other systems and platforms for a seamless experience
  • Explore new use cases and applications for your virtual assistant

Remember, the journey with Sebas is one of continuous improvement and learning. The more you invest in developing your assistant, the more value it can provide to your business and your users.

Conclusion

Creating a virtual assistant with Sebas is an exciting opportunity to revolutionize how your business interacts with customers, automates tasks, and gains insights. With its powerful combination of large language models, customizable prompts, and file attachment capabilities, Sebas puts the power of Generative AI in your hands.

Whether you’re looking to improve customer service, optimize internal workflows, or gain a competitive edge, a well-designed virtual assistant can be a catalyst for growth and efficiency.

As you

begin your journey with Sebas, remember these key principles:

  • Clarity in your system prompt and specific task prompts
  • Selection of an appropriate language model for your needs
  • Adjustment of parameters to achieve the desired performance balance
  • Leveraging file attachments for a specialized knowledge base
  • Continuous iteration and refinement based on real data

With Sebas, you have the opportunity to be at the forefront of the Generative AI revolution and create virtual assistants that truly transform the way you operate. The future is bright, and it starts with the bot you build today.

[Back to Top]