Teaching AI - Knowledge Base Citations v2
One of the many type of citation projects I did where I would evaluate facts within the responses. Then I would evaluate these outputs side by side.
Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
No subject matter listed
Expert in training LLMs in various subjects and tasks such as evaluating prompts and responses, Medical and Biology related domains, STEM related tasks like undergraduate math, physics, and chemistry. Familiarity in various documention such as Markdown and LaTeX. Worked with various LLMs to overcome chatbot hallucinations and data biases. Additionally, I have experience help training various proprietary AIs, such as Claude, ChatGPT, and Google Gemini. I have a strong background in research from my past work as a research assistant and bring forth that expertise into my current AI contract work.
One of the many type of citation projects I did where I would evaluate facts within the responses. Then I would evaluate these outputs side by side.
This project was one of many that involved talking to the LLM about anything that came to mind and evaluate those responses between A or B. I would also edit the responses if they needed to be fact check or written in a better way.
This was a long term project that typically involved various families and subset of projects. We would teach the AI to be more honest, more helpful, and less harmless. We used various methods such as A/B testing of responses, editing responses and prompts, fact checking, citations. I worked on many projects in this domain and even several medical domains and STEM related projects. These involved solving STEM problems step-by-step and getting the model to improve at these subjects. Many of them were high paying projects 40+/hr and included Medical Domain projects of this subset. One involved asking medical related questions and getting the model to correctly cite the responses from medical journals.
Bachelor's in Biology, Biology
Research Assistant