Hybrid AI/ML Architect — Enterprise LLM & Agent Systems

Accenture PLC; Montreal; Hybrid; CAD 109,000 - 220,000; Full time

Job summary

Accenture seeks an AI/ML Architect to design, build, and operationalize AI agents, data pipelines, and ML solutions in regulated environments across cloud platforms.

You will embed with clients to stand up custom agent frameworks across AWS, Azure, and GCP, collaborate with stakeholders to define use cases, prototype agentic workflows, and ship production-ready agents that operate reliably. English communication is required for interaction with stakeholders across Canada, the US, and beyond.

Qualifications

• Bachelor’s degree or college program in a related discipline.
• 2+ years designing and implementing scaled Agentic AI and Generative AI solutions in production.
• 2+ years designing platforms/frameworks used by multiple AI solutions.
• 2+ years building AI/ML solutions on major public clouds using cloud-native services.
• 2+ years with AI tech from public clouds, open source, and 3rd party tools.
• 5+ years Python/Java or equivalent; strong in evaluation tooling, logging and observability.

Responsibilities

• Design AI agents with retrieval, orchestration, tool invocation, evaluation harnesses and observability.
• Build LLM and agentic systems with function-calling and orchestration in production.
• Implement RAG patterns with vector databases and optimize embeddings.
• Train and deploy ML models using PyTorch, TensorFlow, MLflow, SageMaker, Azure ML or Kubeflow.
• Stand up custom agent frameworks across providers and design supporting production architecture.
• Integrate AI agents with enterprise APIs and CI/CD on cloud providers.
• Apply Responsible AI controls in regulated industries.
• Run MLOps/LLMOps pipelines across ML/LLM lifecycle.

Skills

Python

Java

LLM

Agentic AI

MLOps/LLMOps

Education

Bachelor’s degree or equivalent in related discipline

Tools

OpenAI

Claude

Vertex AI

LangGraph

CrewAI

AutoGen

Job description

We are:

The Advanced Technology Centers (ATCs) are the engine for reinvention in our clients’ transformation journey. Powered by more than 255,000 people across 24 countries, ATCs provide our clients with seamless access to industry insights and innovative technology solutions.

We are:

The Advanced Technology Centers (ATCs) are the engine for reinvention in our clients’ transformation journey. Powered by more than 255,000 people across 24 countries, ATCs provide our clients with seamless access to industry insights and innovative technology solutions.

Stronger together!

The Advanced Technology Centers (ATCs) make a tremendous impact in solving our clients’ business problems by leveraging innovation, intelligence, industry insights, new IT, and new technology skills. With the global environment changing at a faster pace, our clients are facing unprecedented challenges, and they need us more than ever before. As a Network, ATCs are positioned to unlock greater opportunities and exponential value for our clients.

The value for our clients and our people:

For our clients, the Network provides the strength of our geographic diversity, greater resilience, and seamless access to the deepest industry knowledge, the latest in Gen AI solutions, and tech expertise from around the world.

For our people, it brings an opportunity to shape truly boundaryless career paths in a highly collaborative team of experts where they can learn from each other and solve the world’s most complex client challenges.

You Are

An AI/ML Architect with strong engineering skills and expertise in LLM and agentic architecture. You design, build, and operationalize AI agents, data pipelines, and ML solutions that run reliably in regulated enterprise environments.

The Work

You embed with clients to stand up custom AI agent frameworks across cloud providers (Amazon Bedrock AgentCore, Azure AI Foundry, Google Vertex AI Agent Builder) and the Anthropic Claude Agent SDK. You partner with stakeholders to define use cases, prototype agentic workflows, and ship production-ready agents that operate reliably.

Responsibilities

Design AI agents with retrieval, orchestration, tool invocation, evaluation harnesses, and lifecycle observability.

Build LLM and agentic systems with function-calling, prompt engineering, and agent orchestration in production.

Implement RAG patterns with vector databases (FAISS, Pinecone, Chroma, Milvus) and tune embeddings for retrieval quality.

Train and deploy ML models with PyTorch, TensorFlow, MLflow, SageMaker, Azure ML, or Kubeflow.

Stand up custom agent frameworks across providers (Amazon Bedrock AgentCore, Azure AI Foundry, Google Vertex AI Agent Builder, Anthropic Claude Agent SDK); design the supporting architecture fo production.

Integrate AI agents with enterprise APIs, microservices, and workflow systems; ship via CI/CD on Azure, AWS, or GCP.

Apply Responsible AI controls (fairness, bias detection, safety filters) in regulated industries.

Run MLOps / LLMOps pipelines for model training, deployment, monitoring, and lifecycle.

This is a hybrid role in Mississauga/Montreal and requires 3 days per week in the office

Here's What You Need:

Bachelor’s degree or completion of a college program in a related discipline

2 years’ experience in designing and implementing scaled Agentic AI and Generative AI solutions that are in operations.

2 years’ experience in designing and implementing Agentic AI and Generative AI platforms and frameworks used by multiple AI solutions.

2 years’ experience in designing, engineering, and operationalizing large-scale AI/ML solutions on at least one major public cloud, using cloud-native AI services and frameworks, open-source technologies, and 3rd party tools.

2 years of experience with key AI technologies from public cloud providers, open source, and 3rd party tooling.

5 years of experience in Python, Java, or equivalent. Comfortable with evaluation tooling, logging, monitoring, and observability.

2 years of hands-on experience with AI platforms (OpenAI, Claude, Vertex AI, open source), agentic orchestration tools (LangGraph, CrewAI, AutoGen), and MLOps/LLMOps practices across the ML and LLM lifecycle.

English is required for this position as this role will regularly interact with stakeholders across Canada, US and other countries across our Global footprint where English is the common language. Due to the significant high volume of interactions with these English-speaking stakeholders, which is inherent to this position, it is not possible to reorganize the company's activities to avoid this requirement.



Compensation at Accenture varies depending on a wide array of factors, which may include but are not limited to the specific office location,

role, skill set, and level of experience. As required by local law, Accenture provides a reasonable range of compensation, based on full-time

employment, for roles that may be hired as set forth below.

The recruiting efforts for this position are intended to fill a brand new position.

The base pay range shown below is intended as a guideline to reflect the majority of offers for this role.

It does not represent a maximum limit — in some cases, actual compensation may exceed the range where appropriate.

Information on benefits is here.

Role Location Annual Salary Range

British Columbia/Ontario/New Brunswick $108,800 to $220,400