HarfangLab is a French cybersecurity scale-up delivering EDR, EPP, and ASM solutions to detect and neutralize modern cyberattacks. Since 2018, we have earned industry recognition, including ANSSI certification in 2020 and ANSSI qualification in 2025, and we are expanding across Europe with a team now numbering 130. We protect the endpoints of more than 800 clients across sectors, including CAC 40 industrials, government entities, SMEs, hospitals, and local authorities.

Role overview: You will join the Artificial Intelligence team (7 engineers) working on a system that leverages a Large Language Model to help cybersecurity analysts detect and respond to threats. The team is expanding to enable autonomous, advanced attack investigations. Responsibilities include: designing, developing, and deploying new features from prototype to production—building investigation agent pipelines; designing an evaluation framework; evaluating foundation models; integrating solutions into the product with CI/CD; deploying and monitoring in production; and potentially working on ML/DL/data engineering projects depending on interests.

About you: Hard skills include Python proficiency; a solid understanding of LLMs (data gathering, prompting, fine-tuning, inference, attention mechanisms, tokenization); hands-on experience with LLMs via APIs or local deployments; and building applications such as RAG systems, tool-calling workflows, agents, or chatbots. Nice to have: basic cybersecurity knowledge; experience with agent frameworks (LangGraph, PydanticAI, Haystack AI); ML/DL frameworks (scikit-lear

HarfangLab is a French cybersecurity scale-up delivering EDR, EPP, and ASM solutions to detect and neutralize modern cyberattacks. Since 2018, we have earned industry recognition, including ANSSI certification in 2020 and ANSSI qualification in 2025, and we are expanding across Europe with a team now numbering 130. We protect the endpoints of more than 800 clients across sectors, including CAC 40 industrials, government entities, SMEs, hospitals, and local authorities.

Role overview: You will join the Artificial Intelligence team (7 engineers) working on a system that leverages a Large Language Model to help cybersecurity analysts detect and respond to threats. The team is expanding to enable autonomous, advanced attack investigations. Responsibilities include: designing, developing, and deploying new features from prototype to production—building investigation agent pipelines; designing an evaluation framework; evaluating foundation models; integrating solutions into the product with CI/CD; deploying and monitoring in production; and potentially working on ML/DL/data engineering projects depending on interests.

About you: Hard skills include Python proficiency; a solid understanding of LLMs (data gathering, prompting, fine-tuning, inference, attention mechanisms, tokenization); hands-on experience with LLMs via APIs or local deployments; and building applications such as RAG systems, tool-calling workflows, agents, or chatbots. Nice to have: basic cybersecurity knowledge; experience with agent frameworks (LangGraph, PydanticAI, Haystack AI); ML/DL frameworks (scikit-lear