Actively recruiting / 59 applicants
We’re here to help you
Cynthia is in direct contact with Viseon AI and can answer any questions you may have. Email
Cynthia, Recruiter About us:
We have an Ecom SaaS product that displays product analytics for e-commerce and dropshipping market analysis.
We want to be able to extend our app to provide users with the ability to generate text descriptions for a product using an LLM. This generated text would predominantly be used for selling the product on their Shopify store. We also want to help them generate Ad copy for social media use for similar reasons.
Requirements:
- (Back-end / Server) Implement an Open-source, Self-hosted LLM model that excels at product detail copy generation and can be hosted cost-effectively. The suggestion would be a model such as Mistral, Llama, DeciLM, or any other conversational model with similar performance. The importance is on speed and efficiency. The LLM should be well supported within the open source community such as on Huggingface and Factor and have low computational overheads when compared with the tokenization cost of a hosted solution like ChatGPT’s API. The task of generating relevant product and ad copy should be considered relatively simple input token length can be considered at approximately 128,000.
- (Front-end) A new page (designs provided) to be added to the web app which takes in a product link from either AliExpress or Amazon passes it to the LLM and returns the generated product description copy or Ad copy (prompts can be provided/aided with)
- (Front-end) Integrate a pop-up modal on our product details page which allows the user to click to open a modal that provides the generated product copy for the current product they are viewing on our app.
Tech Skill Requirements
- Python
- Django
- Postgresql
- Server / Linux
- HTML5 / CSS3
- Scrapy (Python lib)
- Some LLM knowledge