Intermediate LLM Engineer with OpenAI and LangChain experience to support one of our financial services clients- 15284

Intermediate LLM Engineer with OpenAI and LangChain experience to support one of our financial services clients- 15284

Location: Toronto- Hybrid (In-office Tuesday-Thursday; Remote Monday and Friday)

Contract Duration: July 31, 2025 - December 31, 2025

Hours: 9:00 AM to 5:00 PM, Monday-Friday (37.5 hours/week)

Story Behind the Need

• Business group: The role sits within our Canadian Artificial Intelligence Team (AIT), supporting a fast-paced initiative focused on transforming voice call data into actionable insights. The project aims to evaluate and standardize sales call quality, offering deep analysis through cutting-edge AI techniques. To meet accelerating demands, we're seeking a hands-on LLM Engineer-someone deeply technical who writes production-grade code and knows how to optimize infrastructure and APIs for scale.
• This is a high-velocity environment where development cycles move quickly. The ideal candidate thrives in rapid prototyping, iteration, and deployment, balancing research with pragmatic execution.

Description:

• LLM Engineer - OpenAI & LangChain Specialist

The Opportunity

• Join our Canadian Artificial Intelligence Team (AIT) as an LLM Engineer, where you'll lead the development of cutting-edge Generative AI solutions using OpenAI models and the LangChain framework. This role is ideal for someone passionate about building real-world applications with Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and agentic systems. You’ll work at the intersection of AI research and engineering, transforming innovative ideas into scalable, production-ready solutions.

Key Responsibilities

• Design, build, and deploy LLM-powered applications using OpenAI APIs, LangChain, and related frameworks.
• Develop RAG pipelines, agentic workflows, and chatbot systems that integrate with enterprise data sources.
• Fine-tune and optimize LLMs for specific business use cases, ensuring high performance and minimal hallucinations.
• Implement prompt engineering strategies and evaluation frameworks to improve model outputs.
• Integrate LLMs with cloud infrastructure (Azure, Databricks) and tools like Azure Cognitive Search, VectorDBs (e.g., FAISS, Pinecone), and Semantic Kernel.
• Collaborate with cross-functional teams to scope, prototype, and productionize AI solutions.
• Apply LLMOps and MLOps best practices for model deployment, monitoring, and lifecycle management.
• Stay current with advancements in LLMs, GenAI frameworks, and open-source ecosystems.

Candidate Requirements/Must Have Skills:

1. 2+ years of experience in NLP, LLMs and Generative AI.

2. Working knowledge of Model Context Protocol (MCP) and its application in LLM-based systems.

3. Hands-on experience with OpenAI APIs, LangChain, and RAG architectures.

4. Proficiency in Python and related libraries

5. Experience with cloud platform (Azure, AWS, or GCP), especially, Databricks, and Cognitive Search.

6. Familiarity with Vector Databases (e.g., FAISS, Pinecone, Weaviate) and agent frameworks (e.g., Autogen).

7. Solid understanding of prompt engineering, LLM fine-tuning, and evaluation techniques.

8. Experience with DevOps/MLOps tools like Git, Docker, MLflow, and Kubernetes.

9. Strong communication skills and the ability to translate technical concepts into business value.

Nice-To-Have Skills:

1. Experience with Streamlit, Flask, or JavaScript for building interactive frontends.

Education:

• Bachelor's degree in a technical field such as computer science, computer engineering or related field.

Best vs. Average: We value real-world deployment experience. A standout candidate has put LLMs into production, knows how to monitor model behavior post-deployment, and structures code for production-readiness. You’ll be asked about challenges encountered and solutions applied-so familiarity with deployment strategies is key.

This isn't just a hands-off leadership role. You’ll be expected to code, experiment, and build-someone who gets their hands dirty and doesn’t just delegate.

Candidate Review & Selection

•1 round: Virtual interview with technical assessment

•Manager’s Note: Authenticity matters-candidates must not rely on AI tools during the interview

• Hiring Manager’s availability to interview:

See people using AI - make sure candidates are not using AI and reading off a screen

Additional Notes

• Any additional details to share with Suppliers: This project is highly collaborative, involving multiple stakeholders. Strong communication and empathy are bonus-especially when evaluating the quality of human interactions during sales calls.