Lead AI Architect

Kyndryl is looking for a visionary AI Architect to pioneer next-generation AI solutions, integrating Generative AI, Agentic AI architectures, and collaborative Agentic Mesh ecosystems. This role demands expertise in designing self-organizing AI systems that blend large language models (LLMs), small language models (SLMs), and multi-agent frameworks to drive disruptive technological transformation. Candidates must demonstrate mastery in translating theoretical AI advancements into scalable, ethical systems with measurable real-world impact.

Key Responsibilities:

• Advanced Architecture Design:

  • Design Generative AI systems leveraging GANs, VAEs, and transformer-based models for content synthesis, simulation, and data augmentation.
  • Architect Agentic AI frameworks using LangChain, Hugging Face, GCP Agentspace, Azure AI Foundry, AWS Bedrock, and vector databases to enable autonomous problem-solving and adaptive learning.
  • Implement Agentic Mesh ecosystems where AI agents collaborate dynamically, combining language, behavioral, and domain-specific capabilities.

• Model Optimization & Innovation:

  • Develop novel techniques for LLM/SLM fine-tuning, including retrieval-augmented generation (RAG) and context-aware prompt engineering.
  • Optimize inference performance for real-time applications using MLOps pipelines and cloud-native deployment.
  • Lead R&D initiatives in neurosymbolic AI, mechanistic interpretability, and ethical guardrails for autonomous systems.

• Strategic Team Player:

  • Work with high-performance teams in deploying AI systems that redefine industry standards (e.g., healthcare diagnostics, financial forecasting, supply chain resilience).
  • Collaborate on publishing whitepapers and patents on emergent AI architectures, emphasizing sustainability and bias mitigation.

Required Qualifications:

• Education: PhD in Machine Learning, Computational Neuroscience, or AI Ethics (or niche MSc with 8+ years in complex AI systems)
• Technical Expertise:
  • 7+ years building production-grade AI systems using:
    • Generative AI: PyTorch Lightning, Hugging Face Transformers, Stable Diffusion, Open AI, Grok etc.
    • Agentic AI: Hugging Face, Nvidia Toolkits, Hyperscaler AI agent services, LangChain, vector databases, MCP, A2A protocols etc.
    • LLM/SLM Optimization: LoRA, QLoRA, and RLHF techniques.
  • Proven ability to design multi-agent systems with emergent collaboration capabilities.
  • Cognitive & Professional Traits: Lateral thinker with patents/publications in ICML, NeurIPS, or AAAI demonstrating unconventional problem-solving.
  • Portfolio of projects showcasing 10x efficiency gains through AI-driven process reinvention.
  • Fluency in ethical AI frameworks (EU AI Act, OECD AI Principles).

Preferred Qualifications:

• Contributions to open-source AI projects (e.g., LangChain, AutoGPT, Hugging Face).
• Experience with neuromorphic computing or quantum machine learning integration.
• Advisory roles in AI policy committees or standardization bodies.

Language Requirements: Fluency in ethical AI frameworks is expected, with no specific language proficiency mentioned.

What We Offer:

• Opportunity to lead groundbreaking projects at the intersection of Generative AI and Agentic Autonomy with a large base of diverse customers and real-world use cases to solve.
• Access to industry-leading training resources for certifications and accreditations, including Microsoft University and AWS Cloud.
• Employee volunteering and donation platform with access to over 2 million nonprofit organizations.

Join Kyndryl and shape the future of autonomous intelligence systems!