Artificial Intelligence Architect

Appvion; Georgia; On-site; USD 140,000 - 210,000; Full time

Job summary

Appvion is seeking an AI Architect to define the technical foundation for AI/ML across the enterprise, aligning architecture standards, platform choices, and governance with business outcomes. You’ll bridge engineering, data, and strategy to accelerate AI adoption at scale.

You will design reference patterns, evaluate platforms, and lead MLOps practices while mentoring teams and staying ahead of trends to keep our AI roadmap secure, scalable, and compliant.

Qualifications

• 8 years in software or data architecture, with 4 years in ML systems
• Deep expertise in cloud platforms (AWS, Azure, or GCP) and their ML services
• Proven experience designing production ML pipelines at enterprise scale
• Strong understanding of MLOps, model monitoring, and deployment patterns
• Experience with both traditional ML and modern LLM/GenAI architectures
• Familiarity with core enterprise infrastructure architecture

Responsibilities

• Design the enterprise AI/ML architecture, including reference patterns and multi-tenant architectures
• Evaluate and select AI platforms, frameworks, and cloud services
• Establish technical standards for model development, testing, and deployment
• Design agentic search and retrieval systems for enterprise knowledge grounding
• Review and approve architecture for all AI use cases before they reach production
• Define data architecture requirements for ML pipelines
• Lead build vs. buy evaluations for AI tooling
• Mentor technical team members and drive engineering excellence
• Stay current on AI/ML technology trends and assess relevance to roadmap

Skills

Python

Scala

ML architecture

Cloud platforms

MLOps

LLM/GenAI

Data governance

Security by design

Tools

Docker

Kubernetes

CI/CD

MLflow

Model registries

Vector databases

Job description

About The Role

We’re hiring an AI Architect to define the technical foundation for all our AI/ML systems including architecture standards, platform decisions, and quality gates that let us deliver scalable, secure, and governed AI solutions tied directly to business outcomes. You’ll sit at the intersection of engineering, data, and business strategy, designing the systems and setting the standards that accelerate AI adoption across the enterprise.

What You’ll Do

• Design the enterprise AI/ML architecture, including reference patterns and multi-entity / multi-tenant architectures with governed data boundaries
• Evaluate and select AI platforms, frameworks, and cloud services
• Establish technical standards for model development, testing, and deployment
• Design agentic search and retrieval systems for enterprise knowledge grounding
• Review and approve architecture for all AI use cases before they reach production
• Define data architecture requirements for ML pipelines
• Lead build vs. buy evaluations for AI tooling
• Mentor technical team members and drive engineering excellence
• Stay current on AI/ML technology trends and assess their relevance to our roadmap

Qualifications

• 8 years in software or data architecture, with 4 years focused on ML systems
• Deep expertise in cloud platforms (AWS, Azure, or GCP) and their ML services
• Proven experience designing production ML pipelines at enterprise scale
• Strong understanding of MLOps, model monitoring, and deployment patterns
• Experience with both traditional ML and modern LLM/GenAI architectures
• Familiarity with core enterprise infrastructure architecture

Skills

• Languages: Python, SQL, and Scala for ML and data engineering
• ML frameworks: PyTorch, TensorFlow, scikit-learn, and Hugging Face
• MLOps: Docker, Kubernetes, CI/CD, MLflow, and model registries
• Cloud & data: AWS, Azure, GCP, Spark, Airflow, and feature stores
• LLM, GenAI & agentic search: RAG, fine-tuning, prompt engineering, vector databases, query planning, tool use, retrieval orchestration, and multi-step reasoning
• Responsible AI: governance, model monitoring, and security by design
• Solution mindset: design thinking, trade-off analysis, and pragmatic delivery

M2SP