AI modernizer for legacy system migration evidence - ON-1258

Project type: Innovation
Desired discipline(s): Engineering - computer / electrical, Engineering, Computer science, Mathematical Sciences, Business, Social Sciences & Humanities
Company: Agii.ca
Project Length: 4 to 6 months
Preferred start date: As soon as possible.
Language requirement: English
Location(s): ON, Canada
No. of positions: 1
Desired education level: CollegeUndergraduate/BachelorMaster'sPhDPostdoctoral fellowRecent graduate
Open to applicants registered at an institution outside of Canada: No

About the company: 

Agilitas Innovations Inc. (Agii.ca) is a Canada-based technology consulting and AI software company helping organizations modernize legacy systems, automate admin-heavy workflows, and deliver secure digital transformation. Agii combines enterprise program delivery experience with practical software engineering, data analytics, cloud, integration, and AI-agent capabilities. Its current innovation focus is Agii AI Modernizer: a platform that analyzes legacy code and documentation, generates modernization artifacts, maps data and integrations, identifies migration risks, and creates evidence packages for dry-runs, reconciliation, cutover, and rollback. Agii also supports workflow automation, dashboards, reporting, and human-in-the-loop guardrails for regulated or operationally complex environments.

Describe the project.: 

Agii is developing an AI-powered modernization platform that helps organizations understand and de-risk legacy system transformation before committing to large implementation budgets. The project will advance Agii AI Modernizer from a working concept/demo into a repeatable innovation workflow for bounded legacy-code and documentation samples. The platform will ingest source files, technical documents, sample data dictionaries, interface notes, and process descriptions; extract business rules, dependencies, data entities, integration touchpoints, and modernization risks; then generate structured outputs such as reverse-engineering documentation, migration plans, data mapping drafts, risk registers, test/dry-run checklists, and cutover/rollback evidence. The intern will research and prototype retrieval-augmented generation, prompt/workflow orchestration, validation techniques, structured output schemas, and human-in-the-loop review patterns. Methodology will include literature/tool review, architecture design, sample legacy workload preparation, iterative prototype development, evaluation against known reference artifacts, and usability testing with enterprise modernization scenarios. The expected result is a validated prototype and technical playbook showing how AI can accelerate modernization discovery while preserving auditability, traceability, and human review.

Required expertise/skills: 

Required skills: Python or TypeScript/JavaScript; software engineering fundamentals; experience with LLMs, prompt engineering, RAG, vector search, structured outputs, APIs, and data processing; familiarity with cloud deployment, Git/GitHub, databases, and basic cybersecurity/privacy practices. Helpful assets: knowledge of legacy systems such as COBOL, AS/400/RPG, Oracle/Siebel, enterprise integrations, data migration, ETL, system documentation, testing, and audit evidence. The candidate should be comfortable researching technical options, building prototypes, documenting architecture, evaluating output quality, and working in an iterative product-development environment.