At Vocads, we are reshaping customer interaction with our no-code AI voice agents. Our technology automates inbound and outbound calls, boosts team productivity, and delivers a smooth, 24/7 customer experience.
We’re a fast-growing startup, accelerated by Station F and Microsoft GenAI studio, and already trusted by ambitious companies.
Descriptif du poste
Société : Vocads Catégorie : Offre emploi CDI Activité : Informatique Filiere : Assurance Metier : Intégrateur d'applications / Paramétreur ERP Lieu : Paris (Île-de-France) Durée : Indéterminée
Mission
At Vocads, we’re revolutionizing customer interaction through our no-code AI voice agents. Our technology automates inbound and outbound calls, boosts team productivity, and delivers a seamless 24/7 customer experience.
We are a fast-growing startup, accelerated by Station F and Microsoft GenAI Studio, already adopted by ambitious companies and currently expanding into the United States.
LLM DEVELOPMENT AND FINE-TUNING
* Fine-tuning large language models (OpenAI, Anthropic, Meta) for natural and contextual voice interactions.
* Designing NLP pipelines tailored to real-time dialogue and production constraints.
* Customizing models for specific use cases, optimizing for accuracy, consistency, and performance.
* Conducting continuous technology watch to stay up to date with the latest AI advancements.
ADVANCED PROMPTING & REAL-TIME AI CONTENT GENERATION
* Creating and optimizing prompts for precise and contextually appropriate responses.
* Implementing prompt chaining strategies and few-shot learning when needed.
PYTHON DEVELOPMENT, APIS & DEPLOYMENT
* Building and maintaining scripts, notebooks, and APIs to integrate models into our voice applications.
* Using modern Python tools such as Pydantic for data validation and structuring, and Instructor for fine-tuning and embedding optimization.
* Optimizing algorithms and models to enhance accuracy and efficiency.
* Deploying models and pipelines in production, ensuring performance, scalability, and reliability.
* Collaborating with development teams to integrate AI models into existing products and services.
* Working with backend and DevOps teams to ensure continuous monitoring, maintenance, and optimization of models.
PERFORMANCE AND LATENCY OPTIMIZATION
* Analyzing and improving the response times of AI models and pipelines for voice interactions.
* Profiling and tuning models for smooth real-time operation on LiveKit, Twilio, and other voice platforms.
INTEGRATION OF EXTERNAL FEATURES AND DATA SOURCES
* Designing pipelines that allow voice agents to leverage various features or data sources: vector databases, web search, SMS sending, third-party API integrations, etc.
* Developing modular and scalable solutions to enrich model responses based on context and business needs.
* Managing consistency, latency, and security when accessing external data or performing AI-triggered actions.
REAL-TIME MACHINE LEARNING & NLP
* Training, deploying, and optimizing ML and NLP models for low-latency systems.
* Monitoring model performance and continuously improving production pipelines.
Profil recherché
* Degree in Computer Science, Applied Mathematics, AI, or a related field.
* Minimum of 5 years of experience in AI, ML, or NLP, with a strong track record in production and model deployment.
* Expertise in Python and ML/NLP libraries (PyTorch, TensorFlow, HuggingFace, LangChain, etc.), as well as modern tools (Pydantic, Instructor).
* Proven experience in LLM fine-tuning and Prompt Engineering.
* Knowledge of pipelines integrating external functionalities (RAG, APIs, third-party tools, web scraping, SMS, etc.).
* Experience with real-time or interactive systems (latency-critical, performance optimization).
* Ability to design, deploy, and maintain models in production.
* Autonomy, rigor, and a passion for tackling complex technical challenges.
Additional Assets
* Experience with voice platforms (LiveKit, Twilio, WebRTC).
* Experience with MLOps and CI/CD pipelines for ML workflows.
* Familiarity with deployment frameworks (FastAPI, BentoML, TorchServe, etc.).
* Knowledge of cloud environments (GCP).CLIQUER ICI POUR POSTULER
