Principal Architect, AI/ML

As a member of our Cloud Engineering team, you will be working with fast-paced innovative companies, leveraging AI as the key driver of their transformation. Our clients will look to you as their trusted advisor, someone they can rely on and who will be there to help them along their AI journey. You will be expected to cover a large spectrum of technology topics like model optimization, high-performance training on specialized hardware (TPUs), efficient model serving (vLLM), MLOps, and complex agentic systems.

At Zencore, a Principal Architect is a key technical leader in our engineering organization and acts as an ambassador of our technical and cloud engineering expertise. Principal Architects at Zencore are able to navigate a broad technical range but are specialized in one or more domains. They are responsible for the technical oversight and end-to-end delivery of projects within our professional services business.

Whilst this is remote role, we are only considering US based candidates and do not offer visa sponsorship.

What you will do...

• Serve as Zencore’s senior-most technical authority on the practical application of advanced artificial intelligence and machine learning.
• Partner with the sales and business development teams in a pre-sales capacity to scope opportunities, design solutions for proposals, and act as the senior technical voice in client pitches.
• Lead the architecture and design of sophisticated, secure, and scalable AI solutions for our clients, moving beyond standard API integrations to create genuine competitive advantages.
• Collaborate closely with Cloud & Data Architects to guarantee the design and deployment of comprehensive client solutions.
• Address the growing demand for private, data-sovereign AI by designing systems that meet strict GDPR and data privacy requirements. Strive for model explainability and bias mitigation, ensuring solutions adhere to ethical standards and safety guardrails.
• Architect solutions for hosting, fine-tuning, and optimizing both proprietary (e.g., Gemini, Claude) and open-source (e.g., Llama, Mistral) models on hyperscaler platforms.
• Lead clients in selecting optimal cloud-native technologies, prioritizing Google Cloud solutions for deploying and scaling production-grade agentic systems.
• Guide and mentor customers and Zencore's engineering teams on advanced topics, establishing best practices for high-performance training (PyTorch, JAX, TPUs), efficient model serving (vLLM), and complex agentic systems (LangGraph, Langchain, Google ADK).
• Devise the financial architecture of AI solutions by performing ROI analysis and implementing cost-optimization strategies to ensure large-scale deployments remain economically sustainable for customers.
• Act as an external thought leader, contributing to the Zencore brand through blog posts, conference presentations, and community engagement.
• Act as a "player-coach," providing hands-on leadership and fostering a culture of deep technical excellence in AI/ML.

Who we need...

• Master’s degree in Computer Science, natural sciences, mathematics, or a related technical field, or equivalent practical experience in designing and delivering high-scale AI/ML systems.
• Extensive experience in a senior or principal architect role with a proven track record of designing and delivering complex, production-grade machine learning systems that have created measurable business value.
• Deep, hands-on architectural experience with at least one major cloud platform (GCP, AWS, or Azure) is required.
• Direct, hands-on experience with Google Cloud (Vertex AI, GKE, TPUs) is a significant plus.
• Proven expertise in LLM optimization, including techniques for quantization, pruning, efficient fine-tuning (e.g., LoRA), and high-performance serving (e.g., vLLM, TensorRT-LLM).
• Hands-on experience with high-performance ML frameworks (e.g., JAX, PyTorch/XLA) for training or fine-tuning large-scale models.
• Expertise in designing and deploying agentic workflows using both code-centric (e.g., LangGraph, LangChain, Google ADK) and low-code (e.g., Vertex AI Agent Builder, LangSmith Agent Builder) paradigms.
• A strong understanding of the architectural patterns required for building secure, private, and data-sovereign AI solutions.
• Experience with LLM observability and evaluation frameworks (e.g., LangSmith, LangFuse, Vertex AI Evaluation).
• Exceptional communication and stakeholder management skills, with the ability to articulate complex technical concepts and their business value to both technical and non-technical audiences.
• A passion for mentoring and a drive for continuous learning in the fast-evolving AI landscape.