GitHubDiscordIdeas

Core Concepts

Mantium is a powerful data platform that streamlines data retrieval, analysis, and transformation. This Core Concepts page provides an overview of the main components and features of Mantium to help you better understand and utilize the platform.

Suggest Edits

1. Overview of Mantium

Mantium is an innovative data platform designed to simplify data workflows and harness the power of cutting-edge Language Models like OpenAI and Cohere for powerful transformations and enrichments. It enables software developers, data engineers, and data scientists to efficiently connect to various data sources, create and modify datasets, apply transformations, and ship the data for further analysis or storage.

2. Data Sources

Mantium treats data sources as "holy," meaning that the platform never alters the source data, and users are not allowed to modify it either. To apply modifications to your data, you need to create a dataset.

3. Data Sets

Datasets are the primary workspace for applying transformations and enrichments to your data. They allow you to modify and analyze the data retrieved from data sources without affecting the original information.

4. Connectors

Connectors in Mantium are classified into three categories:

  • Sources: Connect to various data sources, such as Slack, Notion, and file uploads, to retrieve data for processing.
  • Destinations: Send processed and enriched data to storage solutions or other platforms for further analysis or visualization.
  • Integrations: Integrate with third-party services, such as Language Models (OpenAI, Cohere) and vector databases (Pinecone), to extend the functionality of Mantium.

5. Transformations

Mantium provides a wide range of transformations to clean, analyze, and restructure your datasets. Transformations include operations like token count, renaming columns, and splitting text, with more options continuously being added.

6. Enrichments

Enrichments are a special type of transformation that typically use AI models, such as OpenAI, to enhance the data with valuable insights and context. Examples of enrichments include text summarization, sentiment analysis, and language translation.

7. Shipping Data

Mantium offers various options for shipping processed data, including:

  • Download: Download the modified data as a file for local storage or further analysis.
  • Vector Databases: Send the data to vector databases like Pinecone for efficient storage and querying.
  • Webhooks: Stream the data to webhooks for real-time processing or visualization.

8. Syncing

Mantium supports syncing data between data sources and datasets, ensuring that your datasets are always up-to-date with the latest information. The platform can automate this process, allowing you to focus on higher-level strategies and decision-making.

By understanding these core concepts, you'll be better equipped to harness the full potential of Mantium and create powerful, data-driven solutions.

Updated 7 months ago

What’s Next

Now that you have a sense of what Mantium can do, lets integrate your application with Mantium.