Data Scientist

Delpath; Toronto; Hybrid; CAD 80,000 - 100,000; Full time

Job summary

A technology firm in Toronto is looking for a candidate with strong expertise in AI and data analysis. The role involves gathering and cleaning data sets for training Large Language Models and developing conversational AI systems. Ideal candidates should have 5-8 years of experience in Python and SQL, along with knowledge in data privacy standards. The position offers a hybrid work model, requiring in-office attendance 2-3 days a week and allows for potential longer-term engagement depending on performance.

Qualifications

• 5-8 years of experience with Python for data analysis.
• 5-8 years of experience with SQL for data manipulation.
• 2-5 years of experience in data preprocessing techniques.
• 2-5 years of experience in data mining and ETL solutions.

Responsibilities

• Gather and clean relevant data sets for training Large Language Models.
• Develop specialized LLMs and advanced NLU techniques.
• Design and optimize intelligent conversational AI systems.
• Ensure strict adherence to data privacy and security standards.

Skills

Python for data analysis

SQL for data manipulation

Data preprocessing techniques

Data mining and ETL solutions

Data privacy and security standards

Job description

Work arrangement

Hybrid - In office 2-3 days per week

Possibility of extension

Depending on performance

Conversion to FTE

Depending on performance

Number of Positions

1

Schedule Hours

9 am-5 pm Monday-Friday (No overtime).

Reason

Extra support

Typical Day in Role

• Determine the "right tool for the job"—balancing the precision of deterministic AI with the flexibility of RAG and AI models.
• Gather and clean relevant data sets for training and fine-tuning Large Language Models (LLMs), and building Retrieval Augmented Generation (RAG) systems.
• Ensure data quality, consistency, and compliance with regulatory requirements.
• Develop and fine-tune specialized Large Language Models (LLMs) and advanced Natural Language Understanding (NLU) techniques for diverse banking use cases, including complex information extraction, multi-turn dialogue management, and financial text summarization.
• Design, develop, and optimize intelligent, agentic conversational AI systems (e.g., advanced virtual assistants, automated financial advisors) leveraging LLMs for seamless customer interactions across various channels, including text and voice.
• Implement and refine advanced Natural Language Understanding (NLU) and Natural Language Generation (NLG) techniques, particularly within Retrieval Augmented Generation (RAG) architectures, to enhance the accuracy, relevance, and explainability of AI responses by grounding them in authoritative data sources.
• Assist in the strategic implementation and deployment of Large Language Models (LLMs) for a wide array of banking applications beyond just customer interfaces, such as automated content generation (e.g., reports, summaries of financial documents), risk assessment insights, and internal knowledge management.
• Analyze interactions with LLM-powered agents and RAG systems to identify emergent patterns, user needs, and critical areas for performance optimization and continuous improvement.
• Generate actionable insights derived from LLM and Agentic AI analysis to strategically inform business decisions and drive innovation.
• Evaluate model performance using relevant metrics and develop strategies for model optimization and bias mitigation.
• Continuously monitor, update, and re-train models to adapt to changing customer behavior, market dynamics, and evolving data landscapes.
• Create dashboards and reporting for a diverse range of stakeholders.
• Ensure strict adherence to data privacy and security standards, especially when handling sensitive customer data within generative AI systems.
• Lead the operationalization of Hybrid Architectures where traditional Intent-based bots meet Generative AI.

Candidate Requirements / Must Have Skills

• 5-8 years of experience with Python for data analysis, modeling, and scripting.
• 5-8 years of experience with SQL for data manipulation and querying.
• 2-5 years of experience in data preprocessing techniques, including data cleaning, transformation, and vectorization.
• 2-5 years of experience in data mining, data profiling, modeling, cleansing, and enriching as well as extracting, transforming, loading (ETL) solutions.
• 2-5 years of experience in data privacy and security standards, especially when handling sensitive customer data within generative AI applications.

Nice-To-Have Skills

• Demonstrable experience or a strong theoretical understanding of Large Language Models (LLMs), including fine-tuning, prompt engineering, and an awareness of their current capabilities and limitations.
• Familiarity with or understanding of Retrieval Augmented Generation (RAG) architectures and their implementation for grounded, factual AI responses.
• Familiarity with Agentic AI frameworks (e.g., LangChain agents, AutoGen) or designing multi-step AI reasoning processes.
• Familiarity with Google Cloud Platform (GCP) or other cloud computing environments.

Best VS. Average Candidate

The best candidate is someone with experience in modelling and data analysis of a contact center/ Voice assistance data sets. Someone with previous experience within financial institutions or similar initiatives.

Candidate review and selection

2 rounds

1st – 2-3 senior managers– 1 hr – MS Teams Video Data science and data analysis experience, one simple coding question (Python preferred ).

2nd – 1-2 senior manager and Director– 30-45 mins – MS Team Video or in-person– situational questions + assess interest in role and cultural fit.