Research Scientist - Large Language Model Post-Training New London, England, United Kingdom · L[...]

PolyAI; London; On-site; GBP 50,000 - 90,000

Job description

Research Scientist - Large Language Model Post-Training

London · Remote - United Kingdom

PolyAI automates customer service through lifelike voice assistants that let customers lead a conversation. Our voice assistants make it possible for businesses to deliver outstanding customer service that rivals their human agents. Our customers, which include the world’s leading logos, are expanding how they use our platform, driving automation of critical customer service operations and integrating PolyAI into their daily customer service workflows.

As a Research Scientist specialising in large language model post-training, you will play a key role in shaping and implementing strategies for aligning language models for use in our conversational AI platform. Your primary focus will be on post-training techniques such as preference-finetuning, reward modelling, synthetic data generation, etc.

Responsibilities:

• Train models and conduct experiments to assess model performance in live deployments
• Develop post-training strategies to achieve state of the art performance on domain-specific tasks
• Generate, collect, and annotate contact center data from sources such as real customer calls, chats, online open datasets, and synthetic data
• Develop robust evaluation benchmarks to track improvements in production models
• Collaborate with the legal and compliance team to address any compliance or data privacy-related issues
• Work closely with product and engineering teams to ensure alignment with business and production goals
• Stay informed about the latest advancements in machine learning, ASR, TTS, and LLM to continuously enhance our technologies

Requirements:

• A degree in Computer Science, Machine Learning, or a related field, or equivalent industry experience
• 3+ years of experience working with deep learning and statistical models
• Strong knowledge of data quality standards and annotation processes, with the ability to independently evaluate and improve models
• Proficiency in Python and familiarity with relevant ML frameworks and libraries (e.g., PyTorch)
• Experience with cloud services such as AWS, GCP, or Azure
• Excellent verbal and written communication skills, with the ability to convey complex technical concepts to diverse audiences
• A passion for solving technical challenges and driving practical solutions

Preferred Qualifications:

• Experience working with LLMs and data preparation pipelines.
• Experience with speech models, such as ASR or TTS.

Benefits:

• Participation in the company’s employee share options plan
• 25 days holiday, plus bank holidays
• Flexible working from home policy
• Work from outside of the UK for up to 6 months each year
• Enhanced parental leave
• Bike2Work scheme
• Annual learning and development allowance
• One-off WFH allowance when you join
• Company-funded fertility and family-forming programmes
• Menopause care programme with Maven
• Private healthcare and dental cover, discounts on gym members and relaxation apps, and access to a range of mental health programs

At PolyAI, we take great pride in our values—they guide everything we do. We believe that a strong culture leads to meaningful work and lasting impact.

Our core values are:

• Only the best: We expect the best from our people, we hire people that expect the best from themselves, and we nurture this drive for excellence.
• Ownership: We care deeply about what we do. We take ownership of our initiatives, decisions, and outcomes.
• Relentlessly improve: We demand more from ourselves and are always evolving. Continuous, obsessive improvement is the only way we will transform the world of conversational AI.
• Bias for action: Our world moves quickly and so do we. We take calculated risks and we deliver impact fast.
• Disagree and commit: We are all working toward the same goal. If we don’t agree with something, we work hard to understand it and when a decision is made, we accept it and give it our all.
• Build for people: We are hyper-focused on delivering the best automated experiences possible so that we can empower people to get exactly what they need, when they need it.

PolyAI is proud to be an equal-opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. All employment decisions at PolyAI will be based on the business needs without attention to ethnicity, religion, sexual orientation, gender identity, family or parental status, national origin, neurodiversity status or disability status.