- Company Name
- Groupe Covéa
- Job Title
- STAGE - Voice_bots avec IA générative
- Job Description
-
Job Title: Intern – Voice Bots with Generative AI
Role Summary: Design, develop, and evaluate generative‑AI voice‑bot solutions for continuous customer engagement, combining large language models, retrieval‑augmented generation, and hybrid rule‑based approaches.
Expectations: Deliver a reusable, adaptable voice‑bot architecture, benchmark models on performance, cost, and deployment ease, and propose scalable enhancements for production use.
Key Responsibilities:
- Review existing natural‑language understanding models and conduct a state‑of‑the‑art survey of voice‑bot techniques (LLM, RAG, supervised models, rule‑based).
- Implement MCP protocols to integrate internal data streams and centralize voice‑bot inputs.
- Build and deploy a generative‑AI voice‑bot, iterating on STT/TTS components.
- Create benchmarking pipelines evaluating accuracy, inference latency, memory consumption, and deployment simplicity.
- Compare multiple model approaches (LLM, classic, rule‑based, hybrid) across defined criteria.
- Propose production‑ready solutions with adjustments for the organization’s environment.
- Collaborate with operational stakeholders to assess impact and feasibility of deployment.
Required Skills:
- Python programming, experience with PyTorch, HuggingFace Transformers, LangChain.
- Knowledge of large language model concepts, evaluation metrics, and RAG/hybrid architectures.
- Familiarity with NLP, Speech‑to‑Text (STT) and Text‑to‑Speech (TTS) technologies.
- Ability to benchmark machine‑learning models on accuracy, latency, memory, and deployability.
- Strong analytical skills, creativity in merging AI with operational constraints, and teamwork orientation.
Required Education & Certifications:
- Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Data Science, or related field.
- Coursework or certifications in NLP, deep learning, and AI model deployment preferred.