Selection of LLM for reading emails with JSON output + Python libraries
Make improvements to the existing application:
1. It is necessary to correctly select the LLM option for reading the mailbox with newsletters from partners
2. Add parsing of images (banners) from Gmail emails in MIME format and reading them using Python libraries. The result of reading the banner will be inserted into the overall prompt for AI
3. Formulate a prompt for the selected LLM model (based on the sample used earlier for another model)
4. Output Json according to the specified sample
Access to the repository with the existing code will be provided, transfer the model for our installation on the server + it will be necessary to write additional code according to the proposed LLM model and the points above in the existing application.
For all questions regarding the already completed application, there will be communication with the developer.
This task is only for those who have already encountered this task.
Client's review of cooperation with Illia S.
Selection of LLM for reading emails with JSON output + Python librariesThe project was not completed, the contractor tried, but it didn't work out.
-
2642 6 1 Hello! I am ready to help with the refinement of your application for integrating LLM with email reading and banner parsing in MIME format. To implement this optimally, it is best to use Python libraries for MIME processing and select LLM considering the specifics of your emails, which will allow for obtaining correct JSON according to the specified template. I have experience working with OpenAI API and customizing prompts for different models, which will help quickly integrate the solution into your current code. I will propose a clear plan for code writing and result verification. I am ready to consider access to the repository and start working soon. I look forward to your questions and details.
-
1616 8 0 Hello,
I am a developer in the AI/ML field. I can complete your project. Write to me, and we will discuss.
-
179 Hello! 👋
I am ready to try to complete your application:
- integration of Gmail API for reading emails in MIME format;
- parsing text and banners from emails;
- text recognition on banners using pytesseract;
- text and banner analysis using LLaMA3;
- formatting the result in JSON.
… 💡 Note: Tesseract language files (~400 MB) are installed on the server, they are not added to the repository.
-
512 3 1 Hello, this can be implemented using LLM Liama Local + RAG, we break photo banners into chunks, embed them and obtain numerical data, after which we will have a response, we also take text from the banners. The prompt will be generated by LLM Liama, how exactly should this be deployed, API, Local command, Telegram Bot, or another interface?
-
6216 74 1 I have experience in image recognition on local models, face transplantation, using small hardware resources. Also experience working with various APIs, including GPT, and many others. Write, we will discuss. I will be happy to help.
-
116 Good afternoon, I will be happy to help with the solution. Please write in private messages.
-
Ви бажаєте серверу модель LLM по типу gpt-4o чи локальну від Liama та інших?
-
локальну
-