Java Solution Architect - AI Native

Role Summary :

Seeking a Java Solution Architect to define and govern scalable, cloud-native architectures for Edward Jones. The ideal candidate brings expert-level Java and Spring expertise, strong architectural leadership, and increasingly, hands-on experience with AI Native architecture patterns including RAG, LLMs, and cloud AI services.

Key Responsibilities :

• Design and own scalable application architectures and microservices for enterprise Java systems.

• Produce High-Level Design (HLD) and Low-Level Design (LLD) documentation.

• Lead architecture reviews and ensure adherence to coding and architectural best practices.

• Mentor and guide development and QE teams on technical decisions.

• Coordinate with onshore (client/partner) teams to align on priorities, requirements, and milestones.

• Design AI Native architectures including RAG-based embeddings, Agentic frameworks, and LLM integrations.

• Work with Azure cloud AI services for AI integration and deployment.

• Implement and manage memory management strategies for AI-powered applications.

• Conduct and lead code reviews across squads.

• Drive DevOps automation and deployment pipelines.

Skills & Experience Required :

• 9+ years of software engineering experience; 13+ years preferred.

• 35 years as an architect on technical design for both on-premises and cloud systems.

• Expert proficiency in Java, Spring Boot, REST APIs, and Spring Batch.

• 3+ years of AI Native architecture experience RAG, LLM deployment, memory management, cloud AI services.

• Hands-on experience with Azure OpenAI, fine-tuning models, and GenAI integrations.

• Strong skills in HLD and LLD, DevOps tooling (Maven, Git, CI/CD, Autosys).

• Good to have : RPM Packaging and Linux.

• Excellent communication with both technical and non-technical stakeholders.

• Subject matter expert across multiple IS disciplines business solutions, systems development, security.

• Note : Candidates with AI Native architecture experience (RAG, LLMs, Agentic frameworks) are strongly preferred.