Agentic AI Developer

Equifax•Toronto, ON

1d•Onsite

About The Position

At Equifax, we are moving past passive AI chat interfaces to build the future of autonomous workflows. We are creating intelligent, self-correcting multi-agent systems that can navigate complex software environments, utilize external tools, and solve open-ended business problems with minimal human intervention. We are seeking a highly skilled Agentic AI Developer to be the core engineer constructing our next-generation AI workforce. In this role, you will be the hands-on builder translating complex automation blueprints into production-grade, stateful agentic workflows. Working within our secure, globally integrated, and highly regulated architecture, you will push the boundaries of what autonomous systems can do—balancing cutting-edge LLM orchestration with absolute data security and deterministic guardrails.

Requirements

3+ years of professional experience building production-grade AI/ML applications, with a heavy emphasis on LLM orchestration and autonomous agent patterns over the last 1–2 years.
Strong proficiency in Python or TypeScript, with a deep understanding of asynchronous programming, API design, and microservices architecture.
Proven, production-level experience building custom agentic runtime loops or utilizing graph-based orchestration frameworks like LangGraph or Google Agent Development Kits (ADK).
Practical experience working with Advanced RAG pipelines, semantic search, and vector databases.
Mastery of JSON schema design and structured decoding techniques for bulletproof LLM tool-calling.
A strong architectural intuition for breaking down ambiguous, multi-step human tasks into structured, programmatic agent prompts and loops.

Nice To Haves

Prior experience deploying AI or high-throughput software solutions within a highly secure or regulated industry.
Experience with deploying live AI solutions with Agentic workflows or GenAI solutions
A strong understanding of the UiPath ecosystem and experience bridging traditional RPA with generative AI models.
A proactive, self-motivated mindset with a passion for driving AI adoption across an organization to fundamentally change the way people work.
Demonstrated learning agility and a proactive approach to mastering new technologies.

Responsibilities

Develop, test, and deploy robust multi-agent architectures and stateful graph workflows using LangGraph, Google ADK frameworks with our internal agentic or UiPath platforms.
Implement advanced short-term and long-term memory systems, utilizing vector databases and custom checkpointing to ensure agents maintain flawless context across long-running, asynchronous tasks.
Apply a customer-centric, design-thinking lens to architecture challenges. You will actively design solutions and rapidly prototype creative, fact-based AI systems that solve ambiguous, non-linear business problems.
Optimize agentic loops for latency, context-window management, and token consumption, making strategic decisions on when to deploy multi-LLM orchestration, lightweight local models, or high-performance frontier LLMs.
Build clean, secure integrations allowing LLMs to interact with internal APIs, databases, modern microservices, and third-party SaaS platforms via advanced function-calling.
Partner directly with infrastructure, product, business and DevOps teams to containerize, deploy, and scale your AI agents securely within a cloud environment, ensuring smooth production delivery.
Partner with our automation squads to integrate agentic decision-making with enterprise-grade UiPath workflows, effectively turning traditional RPA bots into intelligent, cognitive executioners.
Design agent workflows that strictly adhere to enterprise security, compliance, and data governance standards, ensuring auditable decision logs and safe handling of sensitive data.
Build and follow strict software development best practices. You will conduct rigorous code reviews for internal and vendor-delivered artifacts, maintaining a standardized, world-class global code repository.
Build human-in-the-loop (HITL) overrides and strict operational guardrails into agent architectures to eliminate catastrophic hallucinations and prevent infinite execution loops.
Write exceptionally clean, modular, and reusable Python/TypeScript code. Conduct rigorous code reviews for internal and vendor-delivered components to maintain a global standard.
Act as a technical translator, clearly articulating complex AI concepts, loop mechanics, and architectural risks to non-technical business leaders and project teams.