Senior Applied Research Scientist - Voice of the Customer, VOC

The Applied Research Scientist – Voice of the Customer (VOC) - will participate in applied research necessary for the success of Five9’s Voice of the Customer (VOC) AI Agent software platform. Our AI agent platform provides customers with fully autonomous self-service agents for customer support, sales, and services use cases in the contact center. Our Voice of the Customer (VOC) component of this platform analyzes conversations with human agents, and uses them to generate insights, analytics and instructions and operating procedures for AI Agents. In other words – we use our human agent data to train AI agents. This is one of the hottest areas of application for AI Agent technology!

The Applied Research Scientist is responsible for conceiving of algorithms, prompts, clustering and aggregation technologies, agentic systems, and so on, and then running experiments to validate correctness. The applied research scientist will participate in experiment design, execution, results evaluation, documentation, and iteration. They will work on metric definitions and measurement techniques. They will participate in data collection and creation of golden data sets for experiments. They will also run experiments evaluating the performance of different LLMs, including small language models and open weight models.  The applied research scientist will also perform fine tuning of models (typically small language models) and evaluate their performance.

The work of the applied research scientist is directly aligned to the needs of the product itself. Work done by the applied research scientist is meant to quickly make its way into production to improve the product. The applied research scientist is not expected to produce the actual production code, though this can happen in some cases.

Job Requirements

• 5+ years in Conversational AI / related
• 1+ years experience in applied research conversational AI / AI Agents (e.g., multi-agent architecture patterns, prompt engineering and prompt design, context engineering and harness design, REACT, REWOO, Chain of Thought design patterns, memory architectures, and so on)
• Familiarity with one or more AI Agent frameworks, including LangChain, LangGraph, Google and OpenAI ADKs, and so on.
• Software development experience in building simulators and test environments used for running experiments, ideally in python
• Statistical Methodology: Knowledge of advanced techniques and concepts (sampling, A/B testing, regression, properties of distributions, statistical tests and proper usage, etc). Ability to understand model performance
• Business acumen: You understand the bigger picture for customers and the business and know how to probe beyond stakeholders’ stated requests to really understand their needs and the value of information.

Required Qualifications

• Graduate degree in Statistics, Mathematics, Computer Science, Operations Research, Engineering or similar quantitative field.
• 5+ years experience in Python, R, and applied statistics

       This role is hybrid and requires 3 days a week in office.