.svg "Baseline-logo-complet-blanc (9)"){kind=logo}
Digital couponing | Intelligent automation of data capture and processing
.png?width=1500&height=450&name=Banni%C3%A8re%20Baseline%20(23).png "Bannière Baseline (23)")
Client: GoCoupons
Year of Completion: 2022
[Context and problem]
GoCoupons operates a platform that allows its users to access digital coupons and cashback. To receive their refund, customers must submit a photo of their receipt after an eligible purchase. A team member then manually verifies that the items on the receipt match those covered by the coupon and authorizes the rebate when the criteria are met.
This largely manual process results in significant operational costs and hinders the company's growth. Automating it presents a major challenge, particularly because the same product can appear under different descriptions depending on the retailer, and new coupons, associated with new items, are added regularly, requiring frequent system updates.
[Approach]
Our team combined an optical character recognition algorithm on invoice photos with a machine learning algorithm to extract text from invoices and identify relevant items. We found that all stores within the same chain (e.g., Aldo, Sports Experts) use the same labels, significantly reducing the number of possible labels on the invoices. This observation allowed us to develop a learning system based on memorization and text edit distance that quickly adapts to new items and requires minimal training data.
[Results]
The developed solution learns in real time from new item names as invoices are submitted to the GoCoupons platform. After reviewing just 50 invoices, it identifies over 80% of product occurrences with an error rate of less than 5%.
Technology | Strategic identification and prioritization of AI opportunities
.png "Bannière Baseline (18)")
Client: LevelOps
Year of Completion: 2025
[Context and problem]
LevelOps is a Quebec-based SaaS company specializing in optimizing the use of ERP (Enterprise Resource Planning) and MRP (Material Requirements Planning) systems for the manufacturing sector. Its mission is to automate supply chain decision-making through access to data and data-driven business insights.
As a start-up, the company sought to operationalize its expertise and create a relevant service offering, but faced operational challenges, including underutilized management tools, static inventory calculations, and incorrect data that led to duplicates, poor decisions, and financial losses. LevelOps wanted to integrate AI into its platform to enhance its offering and meet its clients' needs while preserving the confidentiality of their sensitive data.
[Approach]
Baseline supported LevelOps through a series of collaborative workshops to clarify priorities and identify high-potential AI use cases. We delivered an AI canvas structuring the identified opportunities, a specifications document detailing the functional and technical requirements, as well as a feasibility test for the integration of Power BI Embedded. Throughout the mandate, we adapted to several strategic shifts, maintaining agile support aligned with the start-up's evolving needs and focused on value creation.
[Results]
This project allowed LevelOps to improve its strategic clarity and confirm its product direction and the technical feasibility of the new product offering. The deliverables supported the prioritization and implementation of a first AI project. The client particularly appreciated our collaborative approach, our ability to adapt to rapid changes in context, and our technical expertise, combined with a strong ability to simplify complex concepts.
Finance | Natural Language Processing for Summarizing and Automated Tasks
.png?width=1500&height=450&name=Banni%C3%A8re%20Baseline%20(21).png "Bannière Baseline (21)")
Client: Laylah
Year of Completion: 2025
[Context and problem]
Laylah is a platform for financial advisors aimed at improving their operational efficiency. The company wanted to integrate AI to use client meeting transcripts to automatically generate summaries and to-do lists for advisors. The objective: to develop a custom solution, hosted in Canada and compliant with legal requirements, while respecting the cost and performance constraints specific to the SaaS (Software as a Service) model.
[Approach]
Baseline supported Laylah through a series of collaborative workshops to define the development plan and analyze technological options that met the sector's legal and technical constraints. We produced detailed product sheets, then contributed to the development by adopting an iterative approach, allowing for rapid adjustments throughout the project.
[Results]
With the development phase successfully completed, Laylah will soon have a new high-performance AI feature and a strategic action plan for its next steps in innovation. The client appreciated our collaboration, our adaptability, and our technical expertise. This project also allowed Baseline to strengthen its experience in AI development with the Scale AI program and in integrating solutions compliant with the strict standards of the financial sector. Laylah will therefore be able to offer a high-performance and competitive solution, capable of generating substantial efficiency gains for its users.
Manufacturing industry | Predictive modeling and model prototyping
.png?width=1500&height=450&name=Banni%C3%A8re%20Baseline%20(24).png "Bannière Baseline (24)")
Client: Bliq Photonique
Year of Completion: 2025
[Context and problem]
Bliq Photonique, a specialist in high-performance microscopes and imaging tools, has developed Cyclop, a probe that measures the viscosity of products in real time directly on production lines. As part of its commercialization, the company wanted to integrate an AI module to transform this data into operational indicators and trigger alerts to optimize cheese cutting, a priority use case that fulfills its mission of improving industrial productivity. However, the analysis conducted by Baseline showed that traditional approaches could not predict the ideal cutting times early enough, justifying the integration of advanced AI models.
[Approach]
Baseline conducted collaborative workshops, analyzed the data collected by the probe, and quickly tested different AI methods. We produced a detailed AI canvas and a specifications document, while validating technical feasibility through a first iteration of models, which allowed us to target the most promising methods for the development phase.
[Results]
Now at the end of its current phase with Scale AI, the project has defined a clear action plan for what comes next: data preprocessing, model training, and integration into the alert system. The client particularly appreciated the clarity of the deliverables and our ability to simplify technical issues. Phase 3 will aim to deliver an operational version, becoming the next key step for the commercialization of the Cyclop probe.
Environment | Content workflow automation and data structuring
.png?width=1500&height=450&name=Banni%C3%A8re%20Baseline%20(25).png "Bannière Baseline (25)")
Client: CPEQ
Year of Completion: 2025
[Context and problem]
The Conseil Patronal de l'Environnement du Québec (CPEQ) is a non-profit organization essential for Quebec organizations concerned about their environmental impact. CPEQ faced a major efficiency challenge in the production of its newsletter. This essential process relied on complex and time-consuming manual manipulations, mobilizing a highly specialized resource for low-value-added tasks. CPEQ therefore approached Baseline with a clear objective: to obtain a ready-to-use solution to automate and optimize the preparation of its newsletter, in order to increase its operational efficiency.
[Approach]
Our approach began with collaborative workshops aimed at thoroughly analyzing CPEQ's existing processes. This detailed understanding allowed us to design a custom automation solution, entirely focused on the tools within the client's existing technological ecosystem. The main technical challenge was carrying out custom integration work with the specific content distribution solution used by CPEQ.
[Results]
The project significantly increased the efficiency of the newsletter's production, thereby freeing up valuable time for the CPEQ team. Beyond this productivity gain, the new solution brought a major additional benefit: it created a structured database of all past mailings. This improvement opens the door to multiple future features that were previously impossible to consider.
CPEQ particularly appreciated our collaborative approach throughout the mandate. The success of this implementation is such that a prioritization of the next features to be implemented is already underway, demonstrating the long-term value and potential of the solution.
Why choose Baseline
Cutting-edge expertise in AI
End-to-end support
Concrete and measurable results
Our AI development method
Developing a high-performing and sustainable AI solution is not something that can be done on the fly. At Baseline, we follow a structured and agile approach that puts you at the heart of the process.
-1.png?width=850&height=1500&name=Banni%C3%A8re%20Baseline%20(7)-1.png "Bannière Baseline (7)-1")
Innovation canvas
Co-construction of the business project to align the technological vision of AI with the strategic objectives of the company.
Statement of Work
Concrete definition of requirements (functional and non-functional), constraints (technical, regulatory) and validation of the sufficiency and accessibility of the data.
Development plan
Breaking down the overall vision into development stages to balance risk management and continuous value delivery.
Prototyping
Rapid validation of the AI-based technical approach, adjustment of the trajectory and construction of a usable prototype serving as the foundation for the final system.
Application development
Building the remaining functionalities around the prototype, using proven iterative methods, and integrating with existing systems (ERP, CRM, etc.).
Deployment
Making functionalities available at scale, solidifying the infrastructure and setting up monitoring and observability tools to detect failures.
Training and knowledge transfer
Training of users and internal teams for the use and maintenance of the solution, ensuring autonomy through clear documentation.
Maintenance évolutive
Post-deployment support to ensure stability, security, adaptability, and to provide updates and integration of new features as needed.