About the role#
The WebHosting Infrastructure team manages the shared infrastructure hosting 6 million customer websites. We are looking for an intern to help develop and evaluate AI agents that pre-analyze production alerts, detect early warning signs, and suggest actions to SREs. This role involves working directly on real production cases to improve operational efficiency.
What you'll need#
- Currently pursuing a bachelor's degree near completion or a master's degree in computer science, software engineering, systems and networks, AI, or a related field.
- Proficiency in Python and familiarity with Linux environments.
- Experience using Git.
- Prior experience with language models or AI agents in an academic or personal project.
- Interest in operations, reliability, and automation.
- Fluent in French and the ability to read technical English.
- Bonus points for knowledge of observability tools, web hosting (Apache, PHP), configuration management (Puppet, Ansible), workflow orchestration (n8n), or prior experience building agents with tool calling.
What you'll do#
- Learn the shared web hosting architecture and monitoring tools.
- Observe SREs during incident triage to understand diagnostic processes.
- Identify high-frequency alerts that consume significant processing time.
- Explore machine learning, statistical methods, and LLM capabilities for log analysis and anomaly detection.
- Create an annotated historical alert dataset for model evaluation.
- Develop reusable tools, such as MCP servers or APIs, to expose diagnostic data like metrics and logs to agents.
- Build and refine an agent that performs automated alert checks and records diagnoses and evidence in tickets.
- Formalize team diagnostic procedures into repeatable skills for the agent.
- Integrate agents into existing automation workflows and extend them to new alert types.
- Design tools to monitor infrastructure trends and flag potential risks.
- Measure agent reliability, team time savings, and incident prevention rates.
- Document technical choices and present regular functional demonstrations.
- Adhere to client data security and confidentiality rules at all times.
Location & details#
- Location: Montréal, Quebec, Canada.
- Duration: 4 to 6 months.
- Modality: On-site.
- This is a full-time, paid internship.


