LLM Research Intern

ProCogia•Vancouver, BC

12h•Onsite

About The Position

We are seeking a motivated and curious Statistical Programming Intern to join our team from July – December 2026. If the co-op is available in July, we would like them to start early and work for part of the summer as well as the defined co-op term from September-December. In this role, you will contribute to the development of internal workflows and tooling that help modernize analytical codebases across SAS, R, and Python, with a particular focus on using AI-assisted development tools to improve translation, testing, and workflow efficiency.

Requirements

Currently enrolled in a university program in third year or above in Applied Statistics, Data Science, Biostatistics, Bioinformatics, Statistics, Computer Science, Mathematics, or a closely related field.
Fluent in R for data work — idiomatic tidyverse to wrangle, join, reshape, and derive variables, with clear reasoning about the statistical operation behind each step.
Comfortable across R’s idioms – vectors, data frames / tibbles, factors, dates, and at least one class from S3/S4/R6. You produce analysis ready tables and graphics.
Comfortable with reproducible analysis — Git for version control, Quarto or R Markdown
Knowledge of SAS is an asset, but we don't expect you to walk in with deep SAS experience. We expect you to be curious about it with a genuine willingness to learn SAS on the job.
Exposure to AI-assisted coding tools (Claude, ChatGPT, GitHub Copilot, Cursor, or similar) and an interest in using them thoughtfully as part of a development workflow.
Some experience with LLMs, prompt engineering, or building automated workflows
Strong problem-solving skills and the ability to work through ambiguity in a hands-on, fast-paced technical environment.
Good written and verbal communication — comfortable explaining your work to teammates, asking questions, and writing things down clearly.

Nice To Haves

Hands on experience with interactive data analysis is plus such as Shiny and plotly
Exposure to clinical data context is a plus — CDISC SDTM/ADaM or pharmaverse packages
Working knowledge of Python is a plus — translation work involves all three languages, and bilingual R+Python candidates ramp fastest.
Hands-on experience with SAS — even at the coursework level. Reading SAS, working with PROC steps and DATA steps, or having translated SAS code before.
Interest in regulated industries (pharma, clinical research, life sciences, financial services) where statistical programming and code quality really matter.

Responsibilities

Contribute to internal workflows and tools that help us translate and modernize analytical code between SAS, R, and Python.
Read and reason about existing SAS programs — understanding the business logic, statistical methods, data transformations, and workflow structure — and help translate them into modern equivalents.
Help design, test, and refine AI-assisted and agentic workflows for code translation, validation, and quality checking. You'll work alongside the team on internal projects first, with the chance to contribute to client-facing work as you ramp up.
Validate that translated code is correct — comparing outputs against the original, checking edge cases, cleaning logs, and documenting how you know the translation is faithful. This validation discipline is the heart of the work.
Work with senior consultants to document patterns, write up translation approaches, and contribute to knowledge-sharing within the team — through demos, internal walkthroughs, or written documentation.