Esperanto is a leader in efficient AI inference specializing in Large Language Models in areas of generative AI. We design the chips, FW, Software, and deliver either fully tested chips, boards, or systems depending on our customer needs.
Interns for the AI team will be working at the Mountain View Office. Interns will be required to be working on site a minimum of 2 days per week for better teamwork and for getting a desirable synergy effect within the team. The duration of the internship should be 2 months or longer, with starting dates and end dates in the May to October window.
Responsibilities:
Speech-to-text project. The intern will evaluate the current state of available pre-trained models in the speech to text area. He will build a system of models which will be able to perform a Speech-to-text summarization. A possible extension of the work would be to look at the vertical application in the medical field where a model would be fine-tuned to assist a physician summarizing his recorded audio of a patient visit.
Minimum Qualifications:
- Advanced Linear Algebra. Python, PyTorch, LINUX. Ability to understand ML technical
papers and present them to the team.
- Grad students will be expected to have studied data sciences/ML and worked on relevant
projects for about 2 years.
- Undergrad students will be expected to already have been exposed to introductory level
Machine Learning and Deep learning classes.
Desired Qualifications:
Statistics. AWS, Parallel Computing, Random processes, Digital Signal processing, NLP, Databases.