Job Description
Join Nexus Horizon Technologies, the frontier of ethical artificial intelligence, as our Lead AI Alignment Engineer. As we approach the era of Artificial General Intelligence (AGI), our mission is to ensure that superintelligent systems remain beneficial, safe, and aligned with human values.
In this pivotal role, you will architect the safety protocols for our next-generation Large Language Models and reinforcement learning agents. You will work directly with top-tier researchers to solve the 'alignment problem' before it scales.
Why Join Us?
We offer competitive equity packages, state-of-the-art compute resources, and the opportunity to define the ethical standard for the next decade of technology.
Responsibilities
- Architect Safety Frameworks: Design and implement robust safety constraints for AI systems to prevent catastrophic misuse and ensure alignment with human intent.
- Red Teaming & Adversarial Testing: Lead adversarial evaluation campaigns to identify vulnerabilities in our models and develop countermeasures.
- Bias Mitigation: Develop and enforce rigorous auditing processes to detect and eliminate algorithmic bias across diverse datasets.
- Interpretability Research: Collaborate with research scientists to build transparent models that explain their internal decision-making processes.
- Policy Development: Translate technical findings into actionable safety policies and guidelines for the broader engineering team.
- Thought Leadership: Publish research papers and present at global conferences on the future of AI safety.
Qualifications
- Education: PhD in Computer Science, Cognitive Science, Philosophy, or a related quantitative field.
- Experience: 5+ years of professional experience in Machine Learning, specifically in Reinforcement Learning (RL), Natural Language Processing (NLP), or Causal Inference.
- Technical Skills: Proficiency in Python, PyTorch, TensorFlow, and familiarity with LLM architectures (e.g., GPT, LLaMA).
- Mindset: Deep understanding of AI safety literature and active engagement with the alignment research community.
- Communication: Exceptional ability to communicate complex technical concepts to both technical and non-technical stakeholders.
- Problem Solving: Demonstrated track record of tackling open-ended, high-stakes problems with innovative solutions.