Guillaume Etchepare

• Core engineer on Tealde: a multi-agent LangGraph system that runs LinkedIn prospecting end-to-end, deployed on AWS Bedrock AgentCore and streaming reasoning, tool calls and interactive widgets to users in real time. • Enabled agents to operate customers' own tools by building a multi-tenant MCP integration layer, with per-connection delegate subagents and per-request tool binding. • Made agent execution safe for production: runtime permission enforcement on tool calls using a taxonomy generated from product manifests, credential scrubbing across logs and traces, and prompt-injection defense on untrusted tool output. • Set up the quality loop for 10+ agents: DeepEval suites built on golden datasets mined from anonymized Langfuse production traces

• Industrialized 5 machine learning and LLM-powered applications, transforming experimental prototypes into production-ready systems • Built and deployed GenAI solutions using LangChain, enabling reliable integration of LLMs into business workflows • Improved the scalability, reliability and maintainability of AI applications through software engineering best practices and large-scale refactoring • Automated testing, deployment and monitoring workflows to accelerate AI product delivery • Containerized and deployed AI services using Docker and Kubernetes • Designed production monitoring and observability systems with Prometheus and Grafana

Member of Multimodal R&D team to build and deploy machine learning models to predict clinical outcomes. Main missions : • Served as technical lead for machine learning development and deployment, mentoring 3 junior data scientists • Developed a reusable Python framework that automated the development, validation and deployment of predictive models • Designed and deployed a custom machine learning platform supporting multimodal healthcare data, including clinical, imaging and genomics datasets • Contributed to peer-reviewed medical research on the application of machine learning to clinical decision-making Based in Boston in 2023/2024.

Consulting position in Microsoft Services to help Microsoft customers to build and deploy AI solutions on Azure. Main contributions : • Delivered AI solutions for leading organizations across manufacturing, automotive and financial services • Scaled machine learning models from prototype to production on Azure cloud environments • Designed and implemented enterprise-grade machine learning platforms and MLOps architectures • Advised customers on MLOps adoption, model lifecycle management and production best practices • Trained more than 50 students on Machine Learning and Azure technologies through Microsoft AI School

Member of IT team to help internal partners (CRM, Marketing, Finance) with implementing and deploying machine learning solutions. Main missions : • Developed predictive models and recommender systems to improve player targeting, engagement and monetization • Built analytics dashboards and data products to analyze the behavior of over 300 million players, supporting profitability, retention and churn analysis • Supervised interns on R&D initiatives focused on next-generation recommendation systems • Designed data models and optimized analytics pipelines to improve ETL performance and reporting efficiency