Our discovery starts with a detailed conversation to determine your highest priority initiatives, for example:
Some companies want to start with a single department or business challenge. Others will look to adopt across the organization from Day 1. Whatever you choose, we will analyze the relevant data you are looking to leverage to determine the best ingestion method. Example datastores may include:
ChatGPT is a conversational Agent built by OpenAI. It has gained recent acclaim for its ability to maintain a conversational dialog, and access a wide array of information on many subjects. Many people interact with ChatGPT via the OpenAI site, but some interact with the model from third party sites or applications.
Fancy Robot similarly allows users to engage with AI models, and can do much more. For example, Fancy Robot can:
Our mission is to create an AI platform built for business, but flexible enough for a multitude of scenarios
During our limited Beta Period, Fancy Robot will be available at no cost for small set of Design Partners.
At the conclusion of the Beta Period, Future Robot will offer a subscription licensed based on usage. Formal pricing has yet to be determined.
Fancy Robot requires only the access necessary to benefit users. For example, if you wished to train your support staff on a large knowledge base, the agent would also need this access. This internal access is what helps Fancy Robot assist businesses at such a large scale.
However, access to specific data is limited on a per-agent basis and only published to internal users with the appropriate privileges.
Chats are a defined as a conversation with an AI model, typically around a single topic. It can be used to ask questions, seek advice, discuss various topics, or simply to have a friendly conversation.
In Fancy Robot, Chats can be saved, shared, organized by folders, and labeled for flexible categorization.
Prompts are defined as the intial message to initiate a chat.
The context and complexity of a promopt can significantly influence the response you receive.
Fancy Robot includes prompt templates for common scenarios, and can leverage variables, conditions, and more.
Agents are software programs that facilitate the interaction between users and language models. The agent receives user requests and forwards them to the appropriate model for processing. Agents can also be used to interact with other data or software if necessary.
Fancy Robot designs purpose-built agents for specific business tasks. These agents are then assigned to appropriate user groups based on their given permissions.
Large Language Models (or LLMs) are deep learning algorithms that have been trained on large amounts of text to understand and generate human language. An example of an LLM is GPT-4.
Fancy Robot will leverage the appropriate models for your use case. Depending on the scenario, Fancy Robot may build logic on top of existing models through the use of Agents, embedded Prompts, or a technique called "fine-tuning", where additional data is added and prioritized over an existing model dataset to access deeper insights on a specific topic.