Job Description
We are building the future of autonomous intelligence. Nexus Horizon is seeking a visionary Agentic AI Lead to spearhead our 2026 Autonomous Systems Initiative. As we transition into the next era of artificial general reasoning, you will architect the next generation of self-correcting, multi-agent AI systems capable of complex, long-horizon planning without human intervention.
In this high-impact role, you will bridge the gap between theoretical research and production-grade engineering, ensuring our AI agents are safe, scalable, and ethically aligned for the year 2026 and beyond.
Responsibilities
- Architect the 2026 Agentic Framework: Design and implement the core architecture for multi-agent systems, focusing on autonomous goal-setting, tool usage, and self-refinement loops.
- Optimize Reasoning Chains: Develop advanced prompting strategies and fine-tuning pipelines to enhance the logical reasoning capabilities of Large Language Models (LLMs).
- System Integration: Integrate autonomous agents with enterprise data lakes, ensuring seamless tool use and external API interactions.
- Mentorship & Strategy: Lead a team of top-tier ML engineers and researchers, setting technical standards and driving the roadmap for autonomous systems.
- Risk Mitigation: Implement rigorous safety guardrails and evaluation metrics to prevent hallucinations and ensure alignment with human values.
- Research Translation: Translate cutting-edge academic papers on reinforcement learning and planning into deployable software solutions.
Qualifications
- Education: Master’s or PhD in Computer Science, Artificial Intelligence, or a related technical field.
- Experience: 5+ years of professional experience in Machine Learning, NLP, or AI Research, with at least 2 years leading technical teams.
- Technical Stack: Deep proficiency in Python, PyTorch, or TensorFlow. Experience with LangChain, LlamaIndex, or similar agent orchestration frameworks.
- Model Expertise: Strong understanding of LLM fine-tuning (PEFT, LoRA) and advanced prompting techniques (Chain of Thought, ReAct).
- System Design: Demonstrated ability to design scalable distributed systems capable of handling high-throughput inference.
- Communication: Exceptional ability to communicate complex technical concepts to non-technical stakeholders and executive leadership.