AI Research Engineer

Diabolocom is seeking an AI Research Engineer to drive the scientific roadmap for AI systems operating on real-world customer conversations. You will investigate novel approaches in NLP and Agentic AI, focusing on challenges like knowledge distillation, few-shot entity extraction, and robust agentic planning, translating research into applied solutions.

• Solve unsolved problems in noise-robust NLP and data-efficient learning. Innovate in data synthesis by building pipelines for training and evaluation. Advance agentic architectures using ReAct, Chain-of-Thought, and tool-use optimization. Design scientific evaluation protocols beyond standard metrics. Stay current with SOTA and contribute to publications.

• Track record of tackling open-ended ML problems. Excellent proficiency in Python. Strong familiarity with deep learning frameworks like PyTorch or TensorFlow.

• Experience with Data-Centric AI (Active Learning, Weak Supervision, Synthetic Data Generation). Advanced tuning techniques (Instruction Tuning, RLHF/DPO, LoRA/QLoRA). Agentic frameworks (LangChain, LangGraph, BAML). Speech processing.