AI Solutions Architect

A national IT consulting company, is seeking an AI Solution Architect to define the architecture and set the technical direction for AI-Native projects. This is a senior, architecture-led role that remains hands-on and credible at the code level, providing technical leadership to the cross-functional engineering team that designs, builds, and deploys production-grade application rewrites using AI-native tools and techniques on the Azure platform.

Responsibilities:

• Own the end-to-end solution architecture for AI-native applications and application rewrites, defining logical components, interfaces, integration patterns, and foundational model selection (LLMs/SLMs).

• Architect and introduce an end-to-end AI-Native SDLC playbook covering the full lifecycle from requirements intake through production deployment, documenting methodology, roles, decision points, trade-offs, lessons learned, and reusable templates.

• Author Architecture Definition Documents (ADDs), reference architectures, and engineering standards (ADRs, runbooks) that guide the delivery team's implementation.

• Translate business objectives into solution designs and validate architectural feasibility through hands-on prototyping, proof-of-concept builds, and reference implementations.

• Build reference implementations using C#, ASP.NET, .NET 10, Python, and Azure with monitoring, logging, and observability, establishing the patterns the engineering team scales to production.

• Architect and build reference RAG pipelines (embeddings, retrieval, re-ranking) and define vector storage solutions (e.g., Azure AI Search, Cosmos DB, pgvector, Qdrant) to meet latency, cost, and quality targets.

• Design and standardize agentic AI frameworks and multi-agent orchestration patterns (e.g., Semantic Kernel, AutoGen, CrewAI, LangGraph, LangChain, LlamaIndex) as reusable orchestration components.

• Integrate LLM and AI capabilities into enterprise applications using Azure OpenAI, OpenAI APIs, and open-source models, progressing reference solutions from prototype to production-ready release.

• Define prompt engineering strategies, prompt versioning, memory management, and task chaining with evaluation coverage using frameworks such as PromptFlow or Prompty.

• Select, configure, and operationalize the AI development toolchain, including IDE integration (e.g., VS Code, Visual Studio), AI coding assistants (e.g., GitHub Copilot, Cursor), agentic development tools (e.g., Copilot Agent Mode), AI-assisted code review, and developer workflow automation.

• Define and implement the release and deployment process, ensuring AI agents and AI-assisted development activities operate within existing enterprise guardrails, including change management, approvals, automated test gates, deployment controls, rollback procedures, and production readiness reviews.

• Integrate application quality, risk, and release-readiness controls into the pipeline, including static analysis, dependency scanning, secret detection, code quality checks, and review gates for AI-generated code.

• Design and govern AI evaluation and quality assurance processes using LLM eval frameworks (e.g., Azure AI Evaluation SDK, DeepEval), including automated regression suites, red-teaming, safety testing, and quality gates for AI-generated outputs.

• Establish AI observability and tracing using Azure Monitor, Application Insights, Dynatrace, LangSmith, and MLflow Tracing (OpenTelemetry) for end-to-end request logging, latency tracking, and trace correlation.

• Ensure solutions adhere to privacy, security, data residency, and ethical-AI regulatory requirements, collaborating with security teams on threat modeling unique to intelligent systems.

• Produce a measurable comparison of the AI-native SDLC against the current lifecycle, covering delivery velocity, defect density, automation coverage, and cost-to-deliver.

• Present pilot outcomes, quantified benefits, risks, and recommendations on scaling the AI-native SDLC methodology to senior stakeholders across future application rewrites.

• Deliver a reusable AI-Native SDLC adoption package, including playbook, reference architecture, toolchain configuration, delivery templates, governance checkpoints, security gates, and production deployment checklist.

• Provide technical direction, coaching, and code review to senior developers and engineers, mentoring the team and raising engineering standards across the portfolio.

Requirements:

• Undergraduate degree in Computer Science or a related STEM (Science, Technology, Engineering or Math) discipline.

• 12+ years of progressive software development and solution/systems architecture experience, including recent hands-on delivery with LLMs or AI integration (an equivalent combination of education and experience may be considered).