About the role#
Join the Public Cloud Infrastructure team to help improve how we manage technical incidents. We are looking for an intern to build a tool that automates the creation of post-mortem reports. You will transform raw data like logs, alerts, and team discussions into structured summaries that outline events, impacts, and corrective actions. This project combines generative AI, automation, and infrastructure reliability.
What you'll do#
- Study methods for automatic summarization, information extraction, and document processing using LLMs.
- Explore Retrieval-Augmented Generation (RAG) techniques to analyze data from multiple sources.
- Develop Python-based connectors to retrieve incident logs, metrics, tickets, and team communications.
- Structure and normalize data to build a coherent incident timeline while filtering out nonessential information.
- Design a prototype that uses LLMs to identify key events and generate structured incident summaries.
- Implement mechanisms to limit AI hallucinations and ensure traceability by linking output to original sources.
- Protect data confidentiality by using internal AI solutions.
- Test the prototype against historical incidents and adjust prompts and extraction methods based on results.
- Document your development process and provide recommendations for potential production deployment.
What you'll need#
- Currently pursuing a master's degree in computer science, software development, artificial intelligence, cloud computing, or a related field.
- Strong programming skills, specifically in Python.
- Experience with LLMs, including prompt engineering and API integration.
- Familiarity with data handling, such as REST APIs, JSON, and timestamped data.
- Comfort working in Linux environments and using the command line.
- Practical experience with Git and software development best practices.
- Ability to read technical documentation in English and write clear summaries.
- Rigor, autonomy, and critical thinking when evaluating AI-generated results.
- Knowledge of OpenStack, networking (TCP/IP, routing), observability tools (OpenSearch, Grafana, Prometheus), or CI/CD practices is a plus.
- Experience with LLM frameworks like LangChain or LlamaIndex is also a plus.
Location & details#
- Location: Montréal, Quebec, Canada.
- Duration: 4 to 6 months.
- Work modality: On-site.
- This is a paid, full-time internship.
About OVHcloud
OVHcloud is a cloud provider based in Roubaix. Founded in 1999, the company operates 43 data centers across four continents. It manages over 450,000 servers to support 1.6 million customers. The organization maintains control over its infrastructure, including server design and fiber-optic network orchestration.
How to get in at OVHcloud
Applying early at OVHcloud gives you a better chance of being seen before the applicant pool becomes unmanageable. Intern Insider sends an instant alert the moment a role matching your target is published, so you can submit your materials among the first. This timing often matters more than you might think when recruiters are sorting through hundreds of resumes. Reaching out to the right people at OVHcloud can also improve your response rates. Intern Insider surfaces the recruiters behind the company roles so you can contact them directly to ask a question or request a referral. A brief, professional note to a human being is usually more effective than waiting for a response from a general portal.


