Data Scientist

About the Role Intermedia is looking for a Data Scientist to join our AI Data Science team. In this role, you will develop and apply advanced machine learning, generative AI, and statistical techniques to improve the intelligence, effectiveness, and reliability of Intermedia's digital agents. You will work on challenging problems involving large language models, retrieval, agent behavior, experimentation, evaluation, and optimization to help create digital agents that can understand customer needs, access relevant knowledge, and successfully complete complex tasks. You will work closely with AI/ML engineers, software engineers, product managers, and other data scientists to take ideas from experimentation through production. You will have significant ownership of data science solutions while helping establish strong practices for measuring and continuously improving the quality of our AI-powered experiences.

What You Will Be Doing

Machine Learning & Agent Intelligence

- Develop, evaluate, and improve machine learning and generative AI solutions that power Intermedia's Digital Agent Platform.

- Apply large language models (LLMs), natural language processing, retrieval, and other advanced AI techniques to improve agent understanding and performance.

- Develop approaches that improve agent reasoning, tool selection, knowledge retrieval, context management, personalization, and task completion.

- Experiment with model, prompt, retrieval, and agent configuration strategies to identify approaches that deliver the best customer and business outcomes.

- Evaluate commercial and open-source models and recommend appropriate approaches based on quality, latency, scalability, and cost.

Generative AI & RAG

- Design, develop, and optimize Retrieval-Augmented Generation (RAG) solutions that ground digital agents in relevant enterprise and customer information.

- Develop and evaluate embeddings, retrieval strategies, ranking approaches, semantic search, and other knowledge-retrieval techniques.

- Build and refine end-to-end pipelines that combine LLMs with retrieval systems, enterprise knowledge sources, and agent workflows.

- Develop approaches to improve the relevance, accuracy, and consistency of AI-generated responses.

- Identify and mitigate issues such as hallucinations, poor retrieval, inappropriate responses, and other failure modes in generative AI applications.

Evaluation & Experimentation

- Develop rigorous evaluation frameworks for measuring digital agent quality, including accuracy, relevance, task completion, reliability, safety, and customer experience.

- Design offline and online experiments to compare models, prompts, retrieval strategies, agent configurations, and other AI approaches.

- Apply statistical analysis, hypothesis testing, segmentation, and other quantitative methods to evaluate AI performance and identify opportunities for improvement.

- Define appropriate metrics and benchmarks that connect model and agent performance to customer and business outcomes.

- Analyze production behavior and feedback to identify patterns, failure modes, and opportunities to continuously improve digital agents.

- Ensure data and model quality through comprehensive testing, validation, and performance evaluation.

Data & Solution Development

- Gather, preprocess, analyze, and model large volumes of structured and unstructured data from multiple sources.

- Use Python, SQL, Spark, and other relevant technologies to build scalable analytical and machine learning solutions.

- Develop features, datasets, and analytical approaches that support model development, experimentation, and evaluation.

- Partner with data and engineering teams to build and maintain reliable pipelines for model training, evaluation, and production use.

- Contribute to scalable ML/AI pipelines that support experimentation, deployment, monitoring, and continuous improvement.

- Ensure solutions are reproducible, maintainable, and designe