Lead Specialist, AI Scientist

This is a fully remote position, open to applicants in Spain.

📋 Description

• Design, develop, and enhance speech language models to improve understanding of spoken responses, pronunciation assessment, fluency, prosody, and communicative effectiveness.

• Create and assess automated scoring and feedback systems for speaking tasks utilized in:

• AI-driven speaking practice offering immediate feedback (learner-focused).

• Job-related oral communication and soft skills evaluations (employer-focused).

• Train, fine-tune, and assess acoustic models, cascading models, speech-to-speech models, speech language models, and scoring models, incorporating neural and large language model methodologies.

• Design experiments and perform quantitative analyses on performance, reliability, and validity to ensure the quality of assessments and integrity of decisions.

• Work across various speaking constructs, such as interactional competence, pragmatic competence, and spoken critical thinking skills.

• Conduct thorough error analysis, intra- and inter-rater reliability studies on ASR outputs, spoken features, and scoring behaviors to inform model and product enhancements.

• Collaborate with product, UX, and assessment scientists to embed models into interactive experiences like practice simulations and hiring processes.

• Implement responsible AI principles in speech systems, ensuring fairness across diverse accents, dialects, and proficiency levels, along with clear feedback and scoring transparency.

• Support the monitoring and governance of models in production settings, ensuring consistent quality and compliance for high-stakes applications.

• Serve as the technical lead for an AI conversational assessment product, closely collaborating with a Product Manager to convert assessment objectives, user requirements, and business constraints into model and system design choices.

• Define the end-to-end design for conversational assessments (task structure, prompts, turn-taking, scoring logic, feedback timing) in partnership with product and assessment stakeholders.

• Weigh assessment validity, user experience, system latency, and scalability when making design trade-offs for production conversational assessments.

⛳️ Requirements

• Master’s degree or PhD in Computer Science, Electrical Engineering, Speech & Language Processing, Applied Linguistics, Language Assessment, or equivalent practical experience.

• Direct experience in building speech recognition, spoken language understanding, or automated scoring systems.

• Proficient programming skills in Python, with experience in PyTorch or similar machine learning frameworks for speech and language modeling.

• Strong foundation in machine learning, statistics, and experimental design, particularly in relation to model evaluation.

• Experience with contemporary neural speech models and large language models, including fine-tuning and assessment for spoken tasks.

• Knowledge of model evaluation metrics pertinent to speech and assessment (accuracy, reliability, validity, fairness).

• Understanding of responsible AI practices, including bias analysis, interpretability, and governance for systems affecting users.

• Excellent communication skills, capable of articulating model behavior and assessment results to both technical and non-technical stakeholders.

• Experience collaborating within cross-functional product teams, contributing to strategic decisions, and deploying machine learning systems into production.

🏝️ Benefits

• Opportunity to work on cutting-edge AI technologies in a collaborative environment.

• Competitive salary and comprehensive benefits package.

• Professional development and training opportunities to enhance your skills.

• Flexible working arrangements to support work-life balance.