Overview In this role you will design, build, and evaluate speech language models for spoken response understanding and automated scoring, supporting learner-facing practice and hiring-facing assessments. You will work with cross-functional teams to deliver accurate, fair, and scalable speech AI that provides real-time feedback and robust evaluation. You will lead end-to-end assessment design and ensure responsible AI practices across accents, dialects, and proficiency levels. This position offers the chance to shape practical, high-stakes language assessment products in production.

ResponsabilidadesDesign and improve speech language models for understanding pronunciation, fluency, prosody, and communicative effectivenessDevelop and evaluate automated scoring and feedback pipelines for speaking tasks (learner-facing and hiring-facing)Train, fine-tune, and evaluate acoustic, cascading, speech-to-speech, speech LMs, and scoring models (neural and LLM-based)Design experiments and analyze performance, reliability, and validity of assessmentsPerform error analysis and reliability studies to guide model and product improvementsCollaborate with product, UX,



and assessment scientists to integrate models into interactive experiencesApply responsible AI to ensure fairness and transparency of feedback and scoresSupport model monitoring and governance in production for high-stakes useAct as technical lead for an AI conversational assessment product and translate goals into design decisionsShape end-to-end conversational assessment design including prompts, turn-taking, scoring logic, and feedback timingBalance validity, user experience, latency, and scalability in production designs Requisitos principalesMaster’s or PhD in Computer Science, Electrical Engineering, Speech & Language Processing, Applied Linguistics, Language Assessment, or equivalent applied experienceHands-on experience building speech recognition, spoken language understanding, or automated scoring systemsStrong Python programming with PyTorch or similar ML frameworks for speech and language modelingSolid grounding in machine learning, statistics, and experimental design as applied to model evaluationExperience with modern neural speech models and large language models, including fine-tuning for spoken tasksExpertise in evaluation metrics for speech and assessment (accuracy, reliability, validity, fairness)Familiarity with responsible AI practices, including bias analysis, interpretability, and governanceStrong communication skills to explain model behavior and assessment outcomes to both technical and non-technical audiencesExperience working in cross-functional product teams and shipping ML systems to productionstrong communicationcross-functional collaborationproblem-solving mindsetPythonPyTorchspeech recognition

📌 Lead Specialist, AI Scientist - Pearson (Madrid)
🏢 Jobrapido
📍 Madrid