תיאור המשרה
Description Curious about what it’s like to work at Cognyte? At Cognyte, you’ll collaborate with expert colleagues around the globe to solve problems most people will never even know exist! You’ll be part of building unique solutions shaped by real investigative methodologies, enabling our customers to identify, investigate, visualize and prevent criminal, terror and security threats worldwide. These solutions are used by law enforcement, national security, and national and military intelligence agencies in almost 100 countries to turn massive, diverse data into clear, actionable intelligence for a safer world. We are looking for an exceptional Data Science Team Lead to spearhead the vision and evolution of our next-gen GenAI Co-Pilot. You will lead a team building a game-changing AI product that revolutionizes how users interact with our software - delivering real-time intelligence, deep investigative insights, and autonomous next-action guidance at scale. As a Cognyter, you will: Define the strategic AI roadmap across the system, serving as the core domain authority for cross-functional initiatives. Architect next-gen capabilities by leveraging cutting-edge LLMs, GenAI, and autonomous agentic workflows to help analysts make critical, real-world decisions. Oversee the end-to-end AI/ML lifecycle, from experimentation and prototyping to deployment, monitoring, and optimization. Partner closely with international product, engineering, and architecture teams to drive high-quality, scalable solutions. Join a team that fears no technology and constantly tests the boundaries of investigative power. Requirements For this mission, you’ll need: M.Sc. or higher in Computer Science, Mathematics, Engineering, or a related quantitative field. 7+ years of hands-on experience building and deploying ML/AI solutions in production, including 2+ years leading or growing a data science/ML team. Proven experience designing and shipping GenAI-based product features and multi-agent systems using modern orchestration frameworks such as LangGraph, CrewAI, or AutoGen, including RAG, vector databases, and tool/function calling. Practical experience with the agentic production stack: observability/eval tooling (e.g., Langfuse), durable workflow orchestration, and typed service design. Strong foundation in classical machine learning — supervised/unsupervised methods, feature engineering, model evaluation — and sound judgment for when to use classical models over LLMs. Advantage: Demonstrated experience fine-tuning and deploying small/large language models, including SFT and preference/RL-based methods such as GRPO or DPO. Advantage: Experience with model optimization and efficient serving: quantization, pruning, LoRA/QLoRA, and high-throughput inference frameworks such as vLLM. Advantage: Experience designing end-to-end ML pipelines (versioning, training/tuning, deployment, testing, monitoring) on cloud-native infrastructure (Kubernetes, GCP/AWS).