Uniphore is one of the largest B2B AI-native companies—decades-proven, built-for-scale and designed for the enterprise. The company drives business outcomes, across multiple industry verticals, and enables the largest global deployments.    Uniphore infuses AI into every part of the enterprise that impacts the customer. We deliver the only multimodal architecture centered on customers that combines Generative AI, Knowledge AI, Emotion AI, workflow automation and a co-pilot to guide you. We understand better than anyone how to capture voice, video and text and how to analyze all types of data.    As AI becomes more powerful, every part of the enterprise that impacts the customer will be disrupted. We believe the future will run on the connective tissue between people, machines and data: all in the service of creating the most human processes and experiences for customers and employees. 
We are currently searching for a bright Deep Learning NLP Engineer for our AI Scientist role to report to our Director of AI Science.This is a tremendous opportunity to work with the best and the brightest in the industry on our world class technologies. 

The ideal candidate will work on

• Building and modifying(fine-tuning) pre-trained enterprise-grade large language models and multi-modal models   
• Able to do a scientific evaluation of NLP/LLM models, come up with new techniques for model validation, evaluation, trust and safety. 
• Able to train/fine tune multi-modal models using RLHF and other reinforcement learning techniques to ensure trust and safety 
• Equipping large language models with the ability tointegrate with other tools 
• Specifying guidelines and designing annotation processes for data, work closely with Data team on data preprocessing, consumption and automated/manual annotations 

Skills & Responsibilities

• Minimum of a Master’s degree including Computer Science, Engineeringor related fields. 
• 3+ years of industry experience in AI, NLP, Vision and ASR 
• In-depth knowledge of deep learning and expertise in PyTorch/TensorFlow. 
• Familiar with Transformers and have used Huggingface library and LLMs to solve NLP problems.
• Worked with large amounts of data and developed approaches to address quality issues, and ability to interpret data with a focus on business goals. 
• Solid expertise in Python programming. 
• Hands-on experience in developing production-ready, scalable code.

Relevant Technologies

• ML: Hugging Face, PyTorch, MLOps tools. 
• Deployment: Docker, AWS, Terraform, Garden 
• Datastores: Postgres, S3. 

Benefits

• Uniphore offers a competitive offer and premium benefits, including insurance, pre-IPO options, leave and other perks. 

Uniphore is an equal opportunity employer committed to diversity in the workplace. We evaluate qualified applicants without regard to race, color, religion, sex, sexual orientation, disability, veteran status, and other protected characteristics. For more information on how Uniphore uses AI to unify—and humanize—every enterprise experience, please visit www.uniphore.com.
All applicants - to review our privacy notice, please click here
Information regarding recruiting and phishing scams: Employment offers will always be made by a Uniphore hiring manager or Talent Acquisition team member with a @uniphore.com email address only. We will never text you about an employment opportunity, interviews or employment offers, and we do not make job offers after only one interview. Additionally, we will never ask you to provide funds for onboarding, supplies or anything else, nor we will ask you to submit your personal information (date of birth, passport details, banking information, social security number, etc.) as part of the interview process. If you are receiving an employment inquiry or employment offer from a non Uniphore.com email address, please assume it is spam. If you believe you have been a victim of a phishing or false employment scam, you may want to report this incidents to the FBI iC3, or visit the Department of Homeland Security’s Cyber Smart website to learn how to report such scam.