Scientist in Residence

The IVADO “Scientist in Residence” program is aimed at startups seeking the scientific expertise of a PhD student in artificial intelligence or data science so as to acquire new skills as well as validate and guide their R&D projects and move them in the right direction.

PhD students taking part in this program acquire hands-on experience in an entrepreneurial environment that is complementary and relevant to their curriculum. This in-company experience enables them to take concrete action in knowledge transfer from academia to industry, and thus contribute to the growth of Québec startups at the leading edge of technological advances in artificial intelligence.

By R&D project, we mean a strategic development project and not a project aimed at rapidly bringing a product or service to market.

Any project involving artificial intelligence and its applications, for example:

• Business intelligence
• Operations research
• Data science
• Data visualization
• NLP
• Machine learning
• Etc.

For examples of mandates, here is a summary of some internship offers from previous editions.

1. Maket

Graph-based Generative Architecture

Maket is a platform that allows architects to facilitate the creation and prototyping of new architectural concepts using artificial intelligence. In a few seconds, Maket enables the architect to generate a wide range of new floor plan concepts based on the client’s needs in a collaborative environment to facilitate communication and presentation of the project to the client. Their clients can easily put annotations on the generated plans and collaborate with the architect in a conceptual phase before  exporting plans into Autocad.

At the moment, Maket allows architects to generate hundreds of floor plans in a few seconds based on a few client criteria. The current result is an image generated via a generative adversarial network (GAN).

Maket would like to develop a complete generative pipeline that adapts to as many constraints as possible to make a floor plan editable and malleable and thus produce plans that exactly meet the architects’ needs. We are currently working on using graph-based approaches that are based on RNN’s, traditionally reserved for molecule generation.

Our goal is to use this technology to improve the efficiency, sustainability and quality of the architectural design process while mitigating the costs and uncertainty that often accumulate during the process.

Specifically, the goal is to be able to direct our research and development team towards plausible hypotheses and maximize our R&D efforts. Knowledge transfer will allow us to test technical hypotheses rapidly. At the moment, our GAN-based output does not allow us to make a plan scalable and does not comprehensively address the physical and regulatory design constraints that the building must have to maximize its efficiency.

Many of our clients have expressed a desire to not only generate an architectural plan using specific constraints, but also to be able to move rooms and expand the project after our algorithm has generated the plan. The architect will be able to define constraints that will be entered into the system and these will be returned in the form of graphics representing a floor plan that respects the constraints provided. If the architect wishes to iterate on a floor plan that the system has produced, they will be able to modify the constraints (add nodes and change node priorities) and then generate a new graph from the base graph taking into consideration the new updated constraints.

Profile required:

• Graph theory
• Machine learning

1. Acrylic Robotics

Machine learning for fine arts

Acrylic Robotics is a robotics startup based in Montreal on a mission to make fine art accessible to the general public. Our technology lets artists make limited-edition painted replicas of original pieces – like a printing press for painted art.

Our motion capture system collects data about each brushstroke and creates a digital fingerprint file that is sent to our painting robots to bring to life, replaying each stroke layer by layer with real paint on canvas. A machine learning-powered optical feedback loop ensures that the robot corrects its mistakes and continuously improves its technique to best match an artist’s style. We’re part of the Next AI Montreal cohort this summer and have previously participated in Centech (Propulsion), Next 36 and the Creative Destruction Lab (AI stream).

Acrylic is looking into the development of a machine learning solution that can accurately estimate the pose of a paint brush that corresponds to an image of a paint stroke. The proposed project will involve the use of reinforcement learning to combine both time series data and imagery to develop a model that can reverse-engineer the motion path of the paintbrush required to create a target brushstroke. This allows us to transition our data collection method away from expensive optical motion capture cameras to normal cameras, allowing us to capture data from artists around the world at a much lower upfront cost.

Profil required:

• Reinforcement learning
• Robotics
• Computer vision
• Movement recognition

1. Greenplay

Automatic detection of transport modes

Greenplay aims to simplify your active or sustainable mobility initiatives by introducing reliable metrics, fun challenges, and tangible rewards. Our mission is to inspire communities to adopt sustainable behaviors through innovative, reliable and easy-to-use technology. Above all, we want to put people and innovation at the heart of sustainable mobility initiatives

In this project, we will explore new ways to improve our mode (bus, train, metro, car, walking, running, cycling and carpooling) detection tools through innovative technologies and new processes. The goal is to work with our team of developers to implement exploratory methods to increase the reliability of the algorithm and processes.

Profile required:

1. Machine learning
2. Statistics and data analysis
3. Interest in gamification
4. Good knowledge of geolocation and sensor data from mobile devices
5. Python and Protocol buffer (GTFS)
6. Tensorflow

Objectives

This program aims to promote the transfer of academic expertise to startups. This is achieved through customized training and transfer activities developed to meet the needs of a startup’s R&D project. In this way, the student gains experience that complements his or her academic curriculum, and the startup benefits from this transfer of knowledge, positioning itself as a scientific leader in its field.

Roles and responsibilities

The customized training will be coordinated by IVADO and delivered by a PhD student specialized in the relevant field, in close collaboration with the startup involved. Following are the roles and responsibilities of the various parties.

IVADO

• Coordination and management of the call for applications issued to startups;
• Receipt, evaluation and selection of applications;
• Communication with and identification of the PhD student interested in the program and the selected R&D projects;
• Administrative and financial management of the program;
• Initial meeting and follow-up meetings during the process to ensure smooth progress and satisfaction of the parties;
• Post-training evaluation and any necessary adjustments to the process.

PhD student / instructor

• Identification of the knowledge transfer goals;
• Identification of the relevant issues to be tackled;
• Identification of and training in available methods and technology tools;
• Validation of the proposed methods and tools;
• Knowledge transfer and support;
• Delivery of the documents developed during the process: e.g. state of the art, documentation on the proposed tools and methods, adoption plan, next steps.

Startup (once selected)

• Identification and mobilization of the R&D team that will lead the project;
• Identification of relevant issues to be tackled during the training and transparency on already tested solutions;
• Availability, transparency and openness to the knowledge transfer to be conducted;
• Before the start of the mandate, payment of a contribution of $2010 plus taxes.

• February 5 to March 10, 2024: call for R&D projects from startups
• Week of March 11: analysis and selection of selected projects
• Week of March 18: 2024: disclosure of results to startups and meeting with selected startups
• From March 18, 2024:
  • communication and sharing with doctoral students
  • selection and administrative agreement
  • payment of contribution by startup and transfer of grant
• From May 1, 2024: start of training and coaching*.

*The start date is determined jointly by the parties.

Knowledge transfer plan

The knowledge transfer plan will cover the 3 components of the program, i.e. preparation and kick-off, training and validation workshops, and transfer. The number of weeks devoted to each stage, as well as the main tasks indicated, may be subject to change depending on the specifics of each project. Following the first stage, a training and transfer plan will be requested.

Given that “Scientist in Residence” is a program for transfer of applied knowledge to an R&D project in artificial intelligence specific to each selected startup, the preparation and startup phase is crucial to achieving smooth progress and the satisfaction of all parties.

We estimate that three weeks (45 h) will be required to complete this first phase. The primary tasks will be structured as follows:

Phase 2: Training and validation workshops

This phase forms the core of the program and, as its name suggests, is the period in which the knowledge transfer will be concretely conducted. Eight weeks (120 h) are set aside for this phase, which will allow time to transfer the theoretical and methodological knowledge as well as the relevant tools for the project, test them in real-world situations, obtain feedback, and make any necessary adjustments.

Phase 3: Transfer and coaching

This last phase consists in ensuring adoption of the best methodologies and tools identified, and completing the knowledge transfer as the R&D project progresses. In addition, the PhD student / instructor will support the startup in defining the next stages of the R&D project. Lastly, this is when the PhD student / instructor finalizes documents benefiting the participating startup at the end of the program (e.g., state of the art, theoretical and practical training, adoption plan). Four weeks (60 h) are planned for this phase.

• The program lasts 15 weeks and must be carried out between May 1 and August 31, 2024.
• The PhD student invests +/- 15 hours per week working with the startup.
• Thanks to the funding provided by ScaleAI, the participating startup will be invoiced by IVADO for just 2010$ + taxes.
• The invoice must be paid on receipt, before the start of the project.
• PhD students will receive a $10,000 grant, managed by their home university.
• A maximum of 15 projects will be funded.

For the startup

• Have less than 20 employees;
• Have an R&D team active in the project submitted;
• Be part of or have completed a program with a recognized entrepreneurial support organization (accelerator or incubator) within the last three years;
• Pay the invoice of $2010 + tx upon receipt to IVADO before the start of the course.

For the PhD student

• Be enrolled in a doctoral program at a Québec university;
• Have the support of his/her thesis supervisor;
• Have no employment or equity ties with the startup.

For PhD students

You can express your interest in participating in the program at any time, simply by writing to entrepreneuriat@ivado.ca.

Your email should include:

• If the program is open, a short paragraph describing your expertise, your CV and the startup you would like to support from the list under the Selected R&D projects tab. We will contact you;
• If the program is not open, a short paragraph describing your expertise, your CV and the sector in which you would like to invest. We will contact you once the projects have been selected.

Once the collaboration with a startup has been confirmed, a complete file must be sent to entrepreneuriat@ivado.ca, including:

• A cover letter outlining your interest in the research project and the relevance of your expertise to the mandate (max. 1 page);
• Your CV in free format;
• A letter of support from your academic supervisor confirming your expertise in the research project and his or her support for your approach. You can use this template.

For the startup

Complete and submit this form before midnight March 10, 2024. Any submission that is incomplete or sent by any means other than the form will be automatically rejected.

— Application period CLOSED (startup only) —

We suggest that you copy and paste your text once written.

If you have any questions, please contact: entrepreneuriat@ivado.ca

A committee will evaluate the projects received:

• Two IVADO technology transfer advisors, including the entrepreneurship advisor;
• One person from the research community;
• the Equity, Diversity and Inclusion (EDI) Advisor.

The committee will evaluate applications and make a selection based primarily on compliance with eligibility criteria, the quality of the proposed project and the project’s impact on the company (number of people benefiting from the transfer). The evaluation committee will be multidisciplinary and in line with EDI principles.

KPI Mining Solutions

Intelligent CapEx Optimization for Sustainable Mining

KPI Mining Solutions (KPI) is a start-up company specializing in AI- and Optimization-based solutions for the mining industry, developed in collaboration with McGill-COSMO Stochastic Mine Planning Laboratory. Our innovative solution, the KPI-COSMO Stochastic Mining Optimizer, is the first commercialized stochastic optimization software for mineral value chains. This technology optimizes production scheduling across the entire value chain, incorporating risk management and leveraging Monte-Carlo-based simulations to minimize geological and financial risk. KPI is a division of KPI Digital, a consulting firm with a 20-year track record in data science, analytics, architecture, AI, and optimization solutions.

The current research initiative at KPI focuses on enhancing the capabilities of the KPI-COSMO Stochastic Mining Optimizer by incorporating capital expenditure (CapEx) decision making into the simultaneous stochastic optimization components. This development aims to increase profitability, integrate expenditure management, improve equipment utilization, and reduce energy consumption. The initiative addresses the critical need for more effective CapEx management in the capital- and risk-intensive mining industry, especially as the demand for critical minerals necessary for a green transition increases. Our AI-based solver includes a smart reinforcement learning (RL)-enhanced hyper-heuristic solver. However, further development is required to create a working prototype for CapEx optimization. CapEx integration is complex due to increased computational complexity and instability. We aim to use advanced machine learning models to develop a meta-approach that uses the solver as a black-box oracle and learns to navigate the CapEx decision space robustly and efficiently.

 Our anticipated areas of support and milestones include:

• conducting a literature review of CapEx optimization tools and combining reinforcement learning and optimization for decision making;
• identifying the most appropriate case study to test and methods to address the problem at hand, balancing computational efficiency, solution quality, and practical applicability;
• effectively transfer knowledge from the Ph.D. student to provide KPI with practical insights to accelerate the integration of CapEx optimization into our products.

Profil required:

• background in reinforcement learning;
• operations research;
• machine learning;
• an interest in stochastic optimization is not required, but would be advantageous.

Want to find out more?

Send your interest and CV to gabrielle.langlois@ivado.ca

V3 Stent

Reinventing the HR function through engagement and AI

V3 Stent, a leader in digital marketing, uses AI to digitalize business, combining advanced technologies and best practices to achieve its customers’ objectives. Its solutions support the growth of SMEs through efficient and cost-effective recruitment (HR), the enhancement of corporate culture (marketing) and the enrichment of customer interactions through AI (sales).

The aim of the project, for which we would like to benefit from a transfer of expertise, is to improve career planning and professional development, and to encourage retention and internal recruitment.

The challenges on which we would like to be supported are as follows:

• creating or adapting a sentiment analytics model for extracting parameters from surveys in order to assess the reference situation for each employee;
• train a career plan recommendation algorithm using a vector space containing all users;
• develop a virtual coach based on the career plan recommendations generated by the similarity algorithm.

Profil required:

• emotion recognition;
• NLP;
• automatic learning;
• recommendation system and vector analytics.

Want to find out more? 

Send your interest and CV to gabrielle.langlois@ivado.ca

____________________

The following projects have already been matched with a lucky student!

Tastet

Curation + Tastet

Tastet is the guide to the best local gourmet addresses.

The “Curation+” project, which is part of the development of the “Tastet+” application, aims to radically transform the restaurant search and evaluation process for the Tastet platform. Currently, this slow and costly manual process limits Tastet’s expansion. “Curation+” aims to harness advances in artificial intelligence to create algorithms capable of correlating vast sets of public data (Google Maps reviews, social networks, reviews, etc.) with in-house evaluation data accumulated over nearly 10 years by Tastet.

The challenges on which we would like to be supported are as follows:

• accurate and reliable interpretation of a wide range of heterogeneous data, especially reviews and text comments. Advances in LLM may offer solutions, but their effective application in this specific context remains to be explored;
• the ability to identify relevant “patterns” in and between public and private data, which would effectively match Tastet’s evaluation criteria.

Profile required:

• UX;
• machine learning;
• LLM.

Want to find out more?

Send your interest and CV to gabrielle.langlois@ivado.ca

Obius

Optimization of predictive planning algorithms for purchasing, replenishment, and inventory management

Obius is a purchasing and replenishment optimization platform for e-commerce and retail businesses. Its solution automates planning decisions and identifies the optimal inventory quantities to order to avoid stockouts or overstocks.

The forecasts of this tool are generated by demand prediction algorithms. Obius wishes to improve the accuracy and efficiency of its algorithms and to integrate machine learning functionalities.

The challenges for which we would like to benefit from a transfer of expertise are:

• improve forecasting algorithms;
• setting up processes and methodologies to increase the accuracy of the algorithms used;
• automate testing, training and selection of algorithms;
• implementation of new, more efficient algorithms.

Profile required:

• demand prediction;
• machine learning;
• deep learning.

Want to find out more?

Send your interest and CV to gabrielle.langlois@ivado.ca

Sentiom

Data analytics and detection algorithm design for intelligent occupant safety solution

Sentiom‘s mission is to digitally transform residences for the elderly and vulnerable into intelligent, sustainable and caring buildings. We’ve developed a deeptech solution using Internet of Things (IoT), 5G, artificial intelligence (AI) and digital twin technologies.

Our project, in clinical partnership with Université Laval’s CIRRIS, aims to harness IoT data from several hundred sensors in our living lab, the Maison Martin Matte. The system collects environmental data such as movement, water consumption, humidity, luminosity, vibrations, barometric pressure, noise level, etc.

The main objective of the mandate is to validate the structure of the data collected, to advise on the architecture of the underlying system, and to develop sophisticated pattern detection algorithms.

The challenges for which we wish to benefit from a transfer of expertise are:

• identifying the preferred methods/algorithms (automations, expert systems, AI) that can be learned in the cloud, but operated with few resources in the fog/edge;
• few anomalies to detect; we have huge datasets of normal observations, but few examples of anomalies;
• third-party labeling; some observations may be mislabeled by customer’s employees on the field;
• data volume management; the abundance of environmental data can lead to storage, processing and latency problems;
• cybersecurity of algorithms; integrity of incoming data, confidentiality of outgoing data, availability of systems.

Expected profile:

• experience in improving model generalization and robustness;
• expertise in statistics, expert systems and machine learning;
• experience with time series or environmental data;
• pragmatic generalist.

Want to find out more?

Send your interest and CV to gabrielle.langlois@ivado.ca

LightX

Medical Revolution through AI – Diagnosis and Monitoring of Eye Pathologies with Stereoscopic Data

LightX is an innovative start-up specializing in the design and manufacture of medical tools and software for ocular diagnostics. The medical tools offered by LightX pave the way for the integration of artificial intelligence tools to support the diagnosis of eye specialists.

LightX develops AI-based ocular diagnostic algorithms from large image databases, bringing with it a number of processing infrastructure and algorithm selection challenges. In addition to diagnostic developments, LightX aims to automate clinical testing, which requires automatic image classification algorithms.

In this context, the challenges for which we would like to benefit from a transfer of expertise are:

• the development and improvement of AI algorithms for the segmentation of eye anatomical structures and the detection of anomalies, notably cataracts and narrow-angle glaucoma;
• integration of these algorithms into the LightX mobile and web application, to offer this diagnostic tool to the various optometrists and ophthalmologists using LightX devices.

Join us in this exciting adventure where you will have the opportunity to contribute to revolutionizing ocular diagnostics and improving visual health services worldwide.

Profile required:

• computer Vision;
• MLOps.

Want to find out more?

Send your interest and CV to gabrielle.langlois@ivado.ca

GroupLabs

Advancing Table-Based Data Retrieval: Integrating Semantic Insights and Correlation-Driven Embeddings with Automated Dependency Mapping

GroupLabs is a R&D and software consulting company that develops a state of the art semantic data integration platform, that provides organizations with a semantic understanding of structured and unstructured data across their entire organization. This allows doctors to retrieve often overlooked relevant medical records, CFOs to understand the weak points in their organizations, or a mine operator to be able to predict a mudslide in the mine saving not only huge costs, but the lives of the people within it. On our consulting side, we do AI CRM consulting, and deliver multiple types of solutions, depending on organization needs. We also integrate our proprietary software into our consulting, allowing users to test our product for free to allow us to iterate and build something people really need.

Our project goal is to integrate semantic insights and correlation-driven embeddings with automated dependency mapping, demonstrating relevant performance gains over comparable systems on the BEIR benchmark. This will be done through the achievement of sub-objectives:

• develop a method that incorporates both correlation information and signal structure analysis into the search process for identifying tables within datasets, aiming to enhance the discovery of valuable and insight-rich data for predictive modeling and analysis;
• Topological Graph Attention Transformers for Unveiling Hidden Structures: Graph Embeddings in the Discovery of Uncharted Relationships in Network Data. Utilizing topological properties for link prediction.

Profil sought:

• graph theory
• ML
• NLP
• signal Analysis
• auto-encoder

Want to find out more? 

Send your interest and CV to gabrielle.langlois@ivado.ca 

Trampoline AI

Unsupervised Domain adaptation and Assessment of document Retrieval in the Enterprise context (DARE)

Trampoline AI offers a unified search solution which connects and indexes all the information sources of a business to provide fast context for internal AI applications and humans. Be it within Google Search, Drive, Gmail, Slack, SharePoint, etc. – with Trampoline, users can always find the information they’re looking for. 

We want you to help us bridge the gap between state-of-the-art research and document retrieval in the enterprise space by enabling metrics-driven development and unsupervised domain adaptation.

Specifically, we’re looking to:

• Identify key metrics for assessing the quality of our document retrieval pipelines and establish pipelines for generating synthetic test sets from an organization’s corpus in an unsupervised (or weakly supervised) way; 
• Establish pipelines for generating training data from an organization’s corpus in an unsupervised (or weakly supervised) way, and use it to train domain-adapted retrieval models (ColBERT or Bi-Encoder) to improve our performance metrics.

You’re a good fit if you have experience in:

• Natural Language Processing;
• Information Retrieval;
• Unsupervised & Weakly Supervised Learning.

Experience or interest in the following areas is a plus:

• Synthetic Dataset Generation with LLMs;
• Contrastive Learning;
• DSPy.

Want to find out more? 

Send your interest and CV to gabrielle.langlois@ivado.ca 

Acrylic Robotics

Robotic Implementation of Painting Machine Learning Model

At Acrylic Robotics we’re dedicated to democratizing the creation and consumption of fine art through our advanced technology. We provide a distinctive service that empowers artists to craft limited-edition painted replicas of their original art pieces, functioning similarly to a contemporary art printing press. Using a specialized drawing software plugin, artists can digitally design their artworks. Our technology meticulously records every aspect of the artwork—such as stroke order, pressure, pigment, and orientation—to generate a digital “fingerprint.” These details are then precisely reproduced on canvas by our painting robots, using top-tier paints to truly capture the artist’s initial vision.

We are currently advancing our machine learning (ML) algorithms to further improve our robotic painting technology. Our goal is to refine the conversion of visual ideas and images into painted artworks, ensuring an accurate representation of the various artistic nuances and styles. This effort is aimed at enhancing our robots’ proficiency in interpreting and implementing complex painting instructions, including the integration of specific structural characteristics and pressure dynamics for more authentic reproductions. Additionally, we are working to streamline the process for generating diverse artistic styles, using continuous machine learning and data version control to fine-tune our model training.

The challenges for which we would like to benefit from a transfer of expertise are:

• increasing the user control on the renderer, allowing for freedom of expression;
• increasing the capabilities of the renderer to conform to different artist styles;
• tuning the ML to robotic pipeline to ensure renders match the painting output.

Profile required:

• base requirements
  • Constrained optimization;
  • Representation learning/engineering;
  • Knowledge of diffusion models.
• nice to have skills
  • Knowledge of bezier curves and splines;
  • AWS + Ops Infrastructure;
  • Experience with .svg.

Want to find out more?

Send your interest and CV to gabrielle.langlois@ivado.ca

As a training and knowledge transfer program, all intellectual property belongs to the startup. If the doctoral student, in conjunction with the company, identifies a new research project and their involvement is desired, then this project must be the subject of a collaborative research agreement separate from this program.

A confidentiality agreement (NDA) may be signed between the doctoral student and the company.

For PhD Student

• Demonstrate integrity and respect in all aspects of your collaboration with the startup;
• Respect the terms of the internship and ensure that objectives and deliverables are met to the best of your ability;
• Mention IVADO in public communications about the project and participate, whenever possible, in IVADO student activities.

For the startup company

• Respect EDI principles when selecting candidates, in accordance with the IVADO EDI reference framework;
• Consult the IVADO brochure on unconscious bias in recruitment;
• Provide a working environment and infrastructure that enables the smooth running and completion of the project;
• Demonstrate integrity and respect in all aspects of your collaboration with the intern;
• Complete the short feedback survey sent by IVADO at the end of the course;
• Mention IVADO in public communications related to work carried out during the program (if applicable).

IVADO’s commitment to equity, diversity and inclusion

To ensure that all members of society benefit equally from the advancement of knowledge and opportunities, IVADO promotes the principles of equity, diversity and inclusion in all its programs. IVADO is committed to providing a recruitment process and research framework that are inclusive, non-discriminatory, open and transparent.