AI insights on proprietary supply chain data - ON-1081

Project type: Innovation
Desired discipline(s): Engineering - other, Engineering, Computer science, Mathematical Sciences
Company: Pitstop Connect
Project Length: 6 months to 1 year
Preferred start date: 09/01/2025
Language requirement: English
Location(s): Toronto, ON, Canada
No. of positions: 1
Desired education level: Master'sPhDPostdoctoral fellowRecent graduate
Open to applicants registered at an institution outside of Canada: No

About the company: 

Pitstop is an AI-powered fleet maintenance platform designed to reduce downtime, optimize maintenance, and improve fleet efficiency. By integrating with telematics, maintenance management systems, and vehicle diagnostics, Pitstop provides predictive insights, automated workflows, and real-time alerts to help fleets stay ahead of potential failures.
Checkout the website pitstopconnect.com

Describe the project.: 

Pitstop is developing an AI-driven query builder that enables fleet managers to generate custom reports tailored to their specific needs. Instead of relying on static dashboards, this innovation will allow users to dynamically extract insights from work orders, preventive maintenance (PMs), DVIRs, and fault data in a structured and meaningful way.

This project will focus on:
1. AI Query Builder for Custom Reports – Leveraging cutting-edge Large Language Models (LLMs) to allow users to input natural language queries and receive structured, reliable reports. This eliminates the need for complex filtering or SQL knowledge, making data more accessible to fleet operators.

2. User-Centric Insights & Actions – Enhancing the dashboard experience by prioritizing recommended actions based on historical work orders, fault trends, and preventive maintenance schedules. The AI will highlight urgent needs and suggest optimized workflows.

3. Predictive Benchmarking & Reporting – Ensuring consistent data outputs by enriching datasets with VMRS classifications, allowing fleets to benchmark repairs and costs based on Make, Model, Year (MMY), mileage, and service history. These insights will help customers proactively adjust maintenance schedules and optimize operational costs.

To achieve practical usability, the system will be designed with LLM-driven query interpretation, structured data enrichment, and continuous accuracy validation. The goal is to ensure consistent, high-quality outputs while maintaining an intuitive user experience.

By integrating machine learning, structured data enrichment, and natural language processing (NLP), this project will transform fleet data into actionable intelligence, making data-driven decision-making seamless and efficient for fleet managers.

Required expertise/skills: 

This project requires expertise in AI/ML development, Large Language Models (LLMs), and structured data processing to ensure reliable and practical fleet insights. Key skills include:

Core Technical Skills:
● Natural Language Processing (NLP) – Experience with LLMs (GPT, Claude, or open-source models like LLaMA) for AI-driven query interpretation.
● Machine Learning & Data Enrichment – Experience applying VMRS coding and improving structured outputs for work order and fault code analysis.
● Backend & API Development – Proficiency in Python, Node.js, and working with RESTful and GraphQL APIs for dynamic data retrieval.
● Database Management – Experience with SQL, NoSQL (MongoDB, PostgreSQL) to optimize structured fleet data storage and querying.
● Data Engineering – Strong understanding of ETL pipelines, data normalization, and accuracy validation methods.

User Experience & Usability Focus:
● Conversational AI Interface Design – Ability to design LLM-driven user interactions that deliver reliable, actionable reports.
● Fleet Operations Knowledge (Preferred) – Familiarity with fleet maintenance workflows, work orders, and DVIRs to ensure practical use cases.

Optional Assets:
● Experience with LLM fine-tuning for domain-specific accuracy.
● Familiarity with fleet telematics systems like Geotab or Samsara.
● Hands-on experience with business intelligence tools (Tableau, Power BI, Looker).