Create a Custom Connector for Prompt Mixer with LLamaIndex Support
Budget: $10 – $30 USD
I am looking for a skilled developer to create a custom connector for the Prompt Mixer, which interacts with the LlamaIndex chat engine. The goal is to build a connector that functions as a chat engine, leveraging models from OpenAI.
1. Connector Development:
- Follow the structure as described in Prompt Mixer’s documentation to create a new connector: Create a Custom Connector Tutorial (https://docs.promptmixer.dev/tutorial-extras/create-a-custom-connector).
2. Integration with LlamaIndex Chat Engine:
- Implement the functionality as outlined in the LlamaIndex documentation for creating a chat engine (https://ts.llamaindex.ai/examples/chat_engine).
- Ensure the interaction can handle conversational queries using the LlamaIndex's `ContextChatEngine` class.
3. Document Loader with SimpleDirectoryReader:
- Set up the document loading mechanism using LlamaIndex's `SimpleDirectoryReader` as described https://ts.llamaindex.ai/modules/data_loader
- The connector should be able to load documents from a specified directory, supporting various file types (.csv, .docx, .html, .md, .pdf).
4. Configuration & Settings:
- The configuration must allow users to specify the path to the directory of documents and their OpenAI API key.
5. Implementation Language & Tools:
- The entire connector should be coded in TypeScript to ensure compatibility and performance.
- Necessary dependencies should be managed via npm or yarn as suitable for the project.
6. Code Examples & Structure:
- For reference on structure and setup, please review these existing connectors:
- https://github.com/PromptMixerDev/prompt-mixer-open-ai-connector
- https://github.com/PromptMixerDev/prompt-mixer-langchain-tavily-search-openai-connector
- https://github.com/PromptMixerDev/prompt-mixer-ollama-connector
1. Connector Development:
- Follow the structure as described in Prompt Mixer’s documentation to create a new connector: Create a Custom Connector Tutorial (https://docs.promptmixer.dev/tutorial-extras/create-a-custom-connector).
2. Integration with LlamaIndex Chat Engine:
- Implement the functionality as outlined in the LlamaIndex documentation for creating a chat engine (https://ts.llamaindex.ai/examples/chat_engine).
- Ensure the interaction can handle conversational queries using the LlamaIndex's `ContextChatEngine` class.
3. Document Loader with SimpleDirectoryReader:
- Set up the document loading mechanism using LlamaIndex's `SimpleDirectoryReader` as described https://ts.llamaindex.ai/modules/data_loader
- The connector should be able to load documents from a specified directory, supporting various file types (.csv, .docx, .html, .md, .pdf).
4. Configuration & Settings:
- The configuration must allow users to specify the path to the directory of documents and their OpenAI API key.
5. Implementation Language & Tools:
- The entire connector should be coded in TypeScript to ensure compatibility and performance.
- Necessary dependencies should be managed via npm or yarn as suitable for the project.
6. Code Examples & Structure:
- For reference on structure and setup, please review these existing connectors:
- https://github.com/PromptMixerDev/prompt-mixer-open-ai-connector
- https://github.com/PromptMixerDev/prompt-mixer-langchain-tavily-search-openai-connector
- https://github.com/PromptMixerDev/prompt-mixer-ollama-connector