Data Scientist

F5 F5 · Enterprise · Dublin, Ireland

Data Scientist specializing in AI/LLMs to build and deploy intelligent tools within a security platform, focusing on AI red-teaming, defense, and securing agentic systems. Responsibilities include designing, developing, fine-tuning, and evaluating LLMs and agents, implementing and deploying these tools, developing novel methods for vulnerability detection, building end-to-end pipelines, and optimizing models for inference.

What you'd actually do

  1. Design, develop, fine-tune, and evaluate LLMs, agents and other deep learning models for AI security tasks (e.g., automated red-teaming operations, vulnerability detection, adversarial attack simulation and intelligent defense mechanisms).
  2. Implement and deploy these AI/LLM-powered tools as robust, scalable features within CalypsoAI's security platform.
  3. Develop and implement novel methods and tools for detecting and mitigating vulnerabilities, feeding directly into the AI Security platform.
  4. Build and maintain end-to-end pipelines optimised for AI/LLM workflows, including data processing, training/fine-tuning, validation, deployment, and monitoring (MLOps for AI).
  5. Optimise AI/LLM models for inference speed, cost, and resource utilisation.

Skills

Required

  • BS or MS in Computer Science, Artificial Intelligence, Machine Learning, or a related quantitative field.
  • 1-3 years of relevant professional experience, including hands-on experience developing, training, or fine tuning AI models, especially LLMs.
  • Experience with AI/LLM libraries (e.g. LangGraph, CrewAI, OpenAI SDK, Pytorch or TensorFlow, Hugging Face Transformers)
  • Understanding of LLM security vulnerabilities (e.g., prompt injection, data poisoning) and mitigation techniques.
  • Proven experience deploying AI/ML models into production environments.
  • Strong analytical and problem-solving skills applied to complex AI systems.

Nice to have

  • Experience specifically fine tuning or training large scale language models.
  • Experience deploying LLMs or generative AI models using frameworks like vLLM, TGI, or cloud provider services.
  • Familiarity with techniques like Reinforcement Learning from Human Feedback (RLHF).

What the JD emphasized

  • hands-on experience developing, training, or fine tuning AI models, especially LLMs
  • Experience with AI/LLM libraries (e.g. LangGraph, CrewAI, OpenAI SDK, Pytorch or TensorFlow, Hugging Face Transformers)
  • Understanding of LLM security vulnerabilities (e.g., prompt injection, data poisoning) and mitigation techniques.
  • Proven experience deploying AI/ML models into production environments.
  • Experience specifically fine tuning or training large scale language models.
  • Experience deploying LLMs or generative AI models using frameworks like vLLM, TGI, or cloud provider services.

Other signals

  • AI red-teaming
  • securing agentic systems
  • operationalise AI models
  • implement and deploy AI/LLM-powered tools