AI Engineer: LLMs, Agents & Automation Lead

purefacts_jobs; Toronto; On-site; CAD 120,000 - 160,000; Full time

Job summary

PureFacts Financial Solutions is seeking an AI Engineer to own the conversational layer and develop a Revenue Assistant agent from R&D to prototype. You will design context architectures around client pricing data and build evaluation and safety frameworks.

The role sits at the intersection of machine learning, software engineering, and product, focusing on building intelligent systems that can reason, automate workflows, and augment human decision-making.

Qualifications

• 2+ years building production, customer-facing LLM/GenAI applications.
• 6-8+ years of total back-end software engineering experience.
• Hands-on with agent frameworks and designing deterministic guardrails in regulated environments.
• Experience in SaaS/fintech or data-driven contexts preferred.

Responsibilities

• Design and build LLM-powered applications and AI agents for internal and client-facing use cases.
• Implement prompt engineering, tool usage, and agent orchestration frameworks.
• Identify opportunities to replace manual processes with AI-driven automation.
• Design RAG pipelines using structured and unstructured data sources.
• Collaborate with data and engineering teams to ensure secure, scalable integrations.

Skills

Python

LLM APIs

Agent frameworks

API & microservices

Data processing

Education

Degree in Computer Science, Engineering, Data Science, or related field

Tools

LangChain

LlamaIndex

LangGraph

OpenAI / Azure OpenAI

Pinecone or Weaviate

AWS / Azure cloud

Job description

PureFacts Financial Solutions is seeking an AI Engineer to own the conversational layer and develop a Revenue Assistant agent from R&D to prototype. You will design context architectures around client pricing data and build evaluation and safety frameworks.

The role sits at the intersection of machine learning, software engineering, and product, focusing on building intelligent systems that can reason, automate workflows, and augment human decision-making.