Patients are frustrated with excessively long wait times and front desk staff are overwhelmed by routine, high-volume calls – driving one of the highest staff turnover rates for any job.

We saw a rare win-win: a way to help both patients and providers at once. So we set out to tackle this problem head on.

Most people know us as the team that developed the first AI for understanding babies’ sounds and voices. We partnered for years with leading children’s hospitals around the world, gathering the largest clinical database of infants’ voices to build our models.

Even after scaling that product to half a million users, the most requested feature was simple: connect me to a provider. I remember thinking: why don’t they just go see their own doctor?

As we dug deeper, it became clear that patients are struggling to get care in a timely fashion. Meanwhile, providers are doing their best and still falling behind.

This should not be the experience that defines Canadian healthcare in 2025.

Working closely with Canada’s AI powerhouses, Mila - Quebec Artificial Intelligence Institute and Vector Institute, we’re now helping providers across the country to deliver the patient experience they always wished they could.

We built Roseline AI, a specialty-specific AI voice agent tailored to the strict safety requirements of healthcare.

Today, when you call your doctor’s office, you’ll likely hear Roseline’s voice, friendly, warm, humanlike. If you need an appointment, she’ll get you one in a minute instead of 30. She’ll give you updates on your prescription refills, and let you know when your labs are ready. She’ll be there regardless of whether you call at midday or midnight.

Roseline is live across Canada, expanding access for tens of thousands of patients. If you’re interested in partnering on this mission, let’s connect!

Shout out to all the forward thinking Canadian physicians, who have been with us on this journey: Ganesh Srinivasan, Niraj Mistry, Dr Eyenimi Ndiomu, Mary Brindle, Luke Kyne, Ilan Shahin, Francis Okwerekwu and many more!

And big cheers to our advisors and investors: Yoshua Bengio, Doina Precup, Jeff Dean, Hugo Larochelle, Marc G. Bellemare, Pieter Abbeel, Radical Ventures, AIX Ventures.

1/ How do I find a good idea to work on? You need a problem that is real and that you care about. One good way to find such a problem is to immerse yourself in that field. E.g. In my case I focused my ML research in pediatric care & on clinical collaborations there which led me to Ubenwa Health.

2/ I have an idea. How do I even start? The first thing on your mind should be how do I test as quickly as possible that my idea can be a business? You need to make an MVP (minimum viable product). In practice, it doesn’t have to be a product, but must answer that question.

3/ After MVP what next? There’s much to learn, in addition to the mindset shift. I recommend joining a good incubator program. You’ll be surrounded by coaches + founders + resources to help you sort out the business preliminaries. In Montreal, two great programs for AI founders are NEXT AI & CDL.

High quality VCs also tend to have great support for their founders. So if you feel you have something special, you could go straight to seeking capital from firms like Radical Ventures & AIX Ventures.

4/ Should I be the CEO or CTO? It should come down to: where will you be most useful to the company? Keep in mind that CEO is a generalist role, in contrast to the specialist mindset that likely brought you to a PhD in the first place.

5/ Should I stay in Canada or incorporate in the US? This should be a business decision. There was a time when US investors required you to have a US-domiciled company. That has changed significantly & investors freely write cheques across both sides of the border.

If you stayed a Canadian corp you also get to benefit from significant non-dilutive funding opportunities. E.g. IRAP, provincial programs, R & D tax credits, etc. Just note that these alone will not make you a successful company.

6/ Should I talk about my idea? Please do. The risk of someone stealing your idea is dramatically lower than the benefits – in most fields. Talking about your idea is an opportunity to dialogue, brainstorm & refine it. Ideas are cheap. It’s execution that makes a great company!

7/ Should I bootstrap or raise VC funding? First thing is there are several options in between these 2 extremes. Eg raising from family and friends, angel investors, family offices, etc. This can definitely be a larger discussion. For now my thoughts are: choose the path that allows you to sleep at night and to effectively pursue your mission.

8/ Can I run my company while doing my PhD? From personal experience I’d say avoid this if you can. Both are intense endeavors that require deep thinking 24/7 and commitment over many years. There will be a cost. But if you’re willing to accept the cost, go for it!

I decided to write this article because since 2019 when I received the Vanier CGS, I have gotten several requests to provide tips and advice, help review, applications, etc. I have had long phone calls, replied emails, LinkedIn chats, etc but it is fast becoming untenable, and I quickly find that I’m repeating and exhausting myself. So with this article I hope to put down in one place some of the things I find most important to think of, some not obvious to the average applicant.

Disclaimer: My views here do not represent that of the Vanier Selection Committee. These are simply my ideas on how you can make a better application based on my own experience as an applicant.

Let’s get to it.

Read the website thoroughly

There is an inclination to just go to the Vanier website download the list of required documents and start preparing them. The application is more than that. There are many nuances to be aware of. Read the full instructions, eligibility, what is expected. Read everything. Read even the instructions to the selection committee - yes it’s on the website.

Three (3) equally-weighted criteria

The 3 criteria (academic excellence, research potential, leadership) upon which applicants are ranked are weighted equally so treat them as such. Don’t say I have 20 publications and a 5.0 CGPA, so I won’t spend time thinking of leadership. Or conversely, don’t say I founded 6 NGOs, so I’m good. Exhaust all possible ways to demonstrate your strength in all 3. Everyone applying is at the top of their game, so be at yours.

Each document has multiple purposes

This was probably the most important advice I received when I applied. It is tempting to think monotonically. E.g. that the research proposal tests your research potential, leadership statement tests your leadership abilities, etc. This is not true. Each document tests multiple criteria. And if you do read the website thoroughly, you will find tables (like the one below) for each criterion. As you can see, the selection committee will be considering your Personal Leadership Statement while assessing you for Academic Excellence. Therefore you should weave in comments relating to this into your statement. If you’re like me, this might not have been obvious from the get-go.

Similarly, the reviewers are asked to find signs of leadership even in your research proposal, and so on. So instead of simply itemising your research accomplishments, make sure to tell a story of you with them. Why did you chose those questions? What critical choices did you have to make in your research career? How are you thinking about broader societal impact of your work? Etc. Note again that this applies to almost every document you will have to provide. When you are done preparing each document, take the tables and ask yourself, does it adequately demonstrate the things the reviewers are looking for?

Start early

Writing is re-writing. As with many things, the earlier you start the more time you have to make it better. You could plan your essay writing in phases. At one point your focus is to mention the key things that should be in there. At another, you want to assess what you’ve written against the criteria tables above. At the latter phase, you want to make the prose flow. You want to make it enjoyable to read, because after all it is humans reading it - make it easy for them to follow your story.

Get all the review you can

Get as much help as you can with reviews - both from domain-specific and non-domain-specific sources. The selection committee will not be experts in your specific field so communicate clearly and use jargon carefully. Most universities offer writing services. Make use of it. Discard the hubris that you’ve been writing English all your life. Get feedback from these people who review hundreds of applications annually. Get feedback from your friends too. There are issues you will never see, that another person will point out to you in a jiffy.

Get essays of previous awardees

Show me and I will remember. You could read tons of articles and listen to advice night and day, but as soon as you read a winning application you realise many things at once. Hopefully you realise how simple it should be, free of clutter, straight to the point, structured, etc. Your university can also help here by giving you access to essays of previous awardees who have consented to share their work. Contact your scholarships unit for this.

If you must, apply again

I applied twice before I got the Vanier, first in 2017, then in 2018. Of course give it your best shot the first time. Many people get it on their first try too. But don’t be discouraged if you don’t get it. If anything you are in a better position the second time ‘cos you have your documents set, you just need to improve them, include the new feathers you have added to your cap and of course chase your professors for reference letters all over again (fun stuff).

That’s it. If there is something you would still like to ask, feel free to email me. I will try to respond to you and also add my response to the FAQs below to help others in the future.

Best,

Charles C Onu (PhD Student, McGill University/Mila)

FAQ

1. Do I need to have publications?

I’d say not neccessarily. Follow the norms in your field. In some disciplines (eg. machine learning these days) it is normal to have 1 or 2 publications during a masters degree. For others, it could be that publishing only comes towards the end of a PhD. In either case, remember that published papers are not the only form of research contributions. Highlight your masters and undergrad theses or other interesting research you did/are doing that is not (yet) published.