AI Recruiting Success Story in Canada: How Syndesus Helped Qingwu Liu, a Computer Vision PhD, Get his Dream AI Job

Breaking into the artificial intelligence industry isn’t always easy, even for candidates with years of highly specialized technical training. Employers may say they need advanced AI expertise, but hiring processes still tend to favor familiar job titles, conventional industry experience, and candidates whose backgrounds fit neatly into predefined categories.

That can create a frustrating gap between what a candidate is capable of doing and what an employer initially sees on a résumé.

Qingwu Liu encountered that gap as he completed his PhD in Montreal and began searching for his first full-time industry position. His academic work involved deep learning, computer vision, object detection, object tracking, transportation safety, and the analysis of video captured by roadside cameras. The technical connection between his research and real-world industrial AI was strong. Nevertheless, when his profile was first presented for a position at Maneva, the initial reaction was that he was a recent graduate without enough professional experience.

The outcome changed because Syndesus did more than forward his résumé.

After learning about Qingwu’s research and understanding how it could translate into a commercial environment, Syndesus recruiter Jaz Guram advocated for him, encouraged the company to reconsider its initial assessment, and helped keep the interview process moving. Within approximately one month of completing his PhD, Qingwu began working as a Forward Deployed Engineer at Maneva, a Toronto-based AI company developing video-to-action technology for manufacturing.

His experience illustrates an important lesson for companies hiring AI engineers in Canada: some of the strongest candidates will not look conventional on paper. Identifying them requires technical understanding, careful candidate evaluation, and a willingness to recognize experience that traditional recruiting processes frequently overlook.

From Vehicle Engineering to Deep Learning and Computer Vision

Qingwu’s path into AI began before he arrived in Canada.

He completed his bachelor’s and master’s studies in China, initially concentrating on vehicle engineering. As autonomous vehicles became a major area of research and development, his work increasingly involved the technologies that allow vehicles and transportation systems to perceive their surroundings, including cameras, sensors, object detection, and tracking.

That interest eventually brought him to Montreal for doctoral research in transportation engineering. Although his PhD was situated within transportation, much of the underlying work involved computer science and artificial intelligence. He collaborated with researchers specializing in computer vision and studied how cameras and deep learning models could be used to identify and track vehicles, pedestrians, cyclists, and other road users.

His doctoral thesis, which he successfully defended at Polytechnique Montréal in May 2026, evaluated deep learning-based object detection and tracking methods for transportation applications. His work examined whether perception systems could reliably support road-safety analysis and pedestrian mobility research.

Qingwu also contributed to research involving road users who rely on mobility aids. The resulting dataset included images of people using wheelchairs, canes, and walkers and was designed to support the development and evaluation of object detection and tracking models.

This was not a candidate who had simply completed several classroom AI projects. He had spent years evaluating deep learning models, working with visual data, studying the reliability of object detection systems, and considering how computer vision performs in complex real-world environments.

The challenge was ensuring that an employer recognized the commercial value of that work.

The Difficulty of Finding a First AI Job After a PhD

Qingwu began his PhD during the first year of the COVID-19 pandemic. Although he entered the program in 2020, he initially studied remotely from China and did not arrive in Montreal until March 2021. Even after arriving, pandemic restrictions meant that he did not begin working regularly in the university laboratory until much later.

That experience affected more than the location from which he conducted his research. It also limited the informal relationships that often help students transition into professional careers.

Graduate students typically build networks through laboratories, conferences, internships, collaborative projects, alumni connections, and frequent interactions with professors and other researchers. Qingwu completed a meaningful portion of his early doctoral work without the normal in-person environment through which many of those connections develop.

By the time he completed the program, he had accumulated approximately 12 years of education across his bachelor’s, master’s, and doctoral studies. Yet he was entering a difficult hiring market as a first-time full-time industry candidate.

This is a common contradiction in AI recruiting. A person may possess unusually deep technical knowledge but still be filtered out because the résumé does not include several years under a recognizable corporate job title. Academic researchers can consequently be treated as entry-level applicants even when they have spent years designing experiments, training models, evaluating results, working with complex datasets, publishing research, and solving problems that directly resemble those faced by AI companies.

Qingwu understood that finding the right opportunity might take time. He had heard of international graduates searching for six months or longer before receiving an offer. He described the technology employment market at the time as effectively “frozen.”

His search changed when he encountered a role being managed by Syndesus.

How Qingwu Connected With Syndesus

Qingwu found the position while reviewing AI job opportunities on LinkedIn.

At that point, he did not know Syndesus or fully understand its role in the hiring process. He simply saw a position that appeared relevant to his background and contacted Jaz Guram, the Syndesus recruiter associated with the opportunity.

Approximately 15 minutes after Qingwu sent the message, Jaz responded. He explained that Syndesus was recruiting on behalf of the employer and proposed an introductory conversation to learn more about Qingwu’s experience.

The two spoke for approximately 50 minutes.

That initial discussion was important because Qingwu’s candidacy could not be fully understood through keywords alone. A résumé might show vehicle engineering, transportation engineering, and doctoral research. A thoughtful conversation revealed how those areas connected through computer vision, camera data, deep learning, object detection, tracking, and real-world perception systems.

After the conversation, Jaz concluded that Qingwu was a strong candidate and prepared to introduce him to Maneva.

This is one of the areas in which specialized AI recruiting differs from general technology recruiting. The recruiter must understand not only what the candidate has done, but how the candidate’s knowledge could transfer into a different technical and commercial context.

Qingwu had conducted computer vision research in transportation. Maneva applies computer vision and video intelligence in manufacturing. The environments were different, but many of the underlying technical problems were closely related.

Why Qingwu’s Research Was Relevant to Industrial AI

Maneva develops video-to-action AI for the factory floor. Its technology is designed to analyze visual information from factory cameras and help manufacturers improve quality, reduce downtime, identify safety concerns, and understand what is happening within production environments.

The relationship between Qingwu’s research and Maneva’s work becomes clear when the technologies are considered at a practical level.

A computer vision model does not inherently care whether an object is a vehicle on a road, a worker on a factory floor, a component moving along a production line, or a product being inspected for defects. The business use case changes, but the foundational challenges may remain similar:

• collecting and preparing visual data;
• detecting relevant objects or activities;
• tracking movement across frames;
• evaluating model accuracy and reliability;
• adapting a system to a particular physical environment; and
• translating model output into information that a customer can use.

Qingwu had developed expertise in several of these areas through his doctoral research. He understood how camera-based data is processed and how detection and tracking systems behave outside controlled demonstrations. He had also studied the limitations of these models, which is especially valuable when AI must perform reliably in operating environments rather than only in laboratory conditions.

His academic specialization therefore gave him a relevant foundation for industrial computer vision, even though his previous work was categorized as transportation research.

Recognizing that transferability was the first important contribution Syndesus made to the hiring process. The next was persuading the employer to consider it.

When the Initial Response Was “Not Enough Candidate Experience”

The first response from the employer was not an interview invitation.

According to Qingwu, the initial assessment was that he had just graduated and did not possess sufficient work experience. A conventional recruiting process might have ended at that point. The recruiter could have accepted the response, closed the application, and moved to the next résumé.

Instead, Jaz pushed back.

He explained that Qingwu’s PhD should not be treated as an absence of experience. During the program, Qingwu had completed substantial technical work directly relevant to the position. Jaz encouraged the company to speak with him before deciding whether his background was suitable.

The employer agreed to conduct an interview.

That intervention was consequential. Syndesus did not ask the company to lower its technical standards or overlook a missing qualification. It asked the company to evaluate the candidate’s actual experience rather than relying on a narrow definition of where relevant experience can be acquired.

For employers, this distinction matters. An AI PhD should not automatically be assumed to possess production experience, customer-facing ability, or commercial judgment. Those capabilities still need to be assessed. However, it is equally inaccurate to assume that years of applied research amount to no experience at all.

The purpose of a strong recruiting partner is to make that distinction carefully and give qualified candidates an opportunity to demonstrate what they can do.

A Fast Three-Round AI Interview Process

Once Maneva agreed to meet Qingwu, the process moved quickly.

He completed three interviews: an initial conversation with human resources, a technical and leadership discussion with the chief technology officer, and a final interview with a project manager in Quebec.

The final conversation introduced another dimension to the role. Because the position involved supporting customers in Quebec, the company needed someone capable of communicating in both English and French. Qingwu had learned French while living in Montreal and was able to complete that portion of the interview in French, even though he did not consider himself perfectly fluent.

The CTO interview was arranged particularly quickly. After Qingwu completed his first interview, Jaz contacted him about the next stage, which could occur almost immediately. That kind of fast coordination supports candidate experience and shortens time to hire, but strong AI recruiting should support data-driven hiring decisions without replacing human judgment. Qingwu was nervous because it was his first time interviewing directly with a technology executive and he had little time to prepare.

Jaz reminded him that they had already discussed his experience in depth. Qingwu understood the technical work and did not need to manufacture a different version of himself for the interview. The guidance was simple: rely on the knowledge developed during his master’s and PhD studies, answer naturally, and be comfortable acknowledging that there might be questions he could not answer.

For Qingwu, “just be yourself” turned out to be the most useful interview advice.

Throughout the process, Jaz remained in contact with both sides. He followed up with the company, asked about the status of each stage, communicated with Qingwu, and helped prevent the process from losing momentum.

Qingwu felt that the interviews moved faster than processes he had experienced with other companies. He described Jaz as being on his side and helping move the company toward a decision.

Becoming a Forward Deployed Engineer at Maneva

The process ended with an offer for Qingwu to join Maneva as a Forward Deployed Engineer, a placement he publicly credited Jaz and Syndesus with helping him secure.

The role combines AI engineering with direct customer engagement.

Qingwu may spend a significant portion of his time at customer factories, helping install or configure camera systems, understanding the customer’s environment, and collecting the visual data required for a particular use case. That data can then be brought back to the technical team, used to train or refine models, and ultimately transformed into a working solution that can be deployed for the customer.

The position therefore sits at the intersection of several capabilities:

• computer vision and deep learning knowledge;
• an understanding of camera-based data;
• model training and evaluation;
• field deployment;
• communication with customers;
• problem-solving in physical operating environments; and
• the ability to explain how an AI product should be used.

For Qingwu, it represents a natural progression from studying how computer vision systems interpret activity on roads to helping AI systems interpret activity inside factories.

It is also the kind of AI position that can be particularly difficult to fill through keyword-based recruiting. The ideal candidate is not only a machine learning engineer. The employer also needs someone who can travel to customer sites, understand physical systems, communicate across technical and nontechnical teams, and operate comfortably in real-world deployment environments. Effective candidate matching depends on going beyond a generic job description to assess skills in context and improve candidate quality.

The Result: A First AI Role Within a Month of Graduation

Qingwu estimated that he submitted approximately 30 to 40 applications during his job search. Roughly one month after completing his PhD, he began his first full-time industry position, a pace that points to shorter time to hire and stronger recruiter productivity when the process is focused.

The relatively fast result was significant given the experiences of other international graduates he knew. Some had spent many months applying without securing a role, despite holding advanced degrees and valuable technical skills. It also reflects AI’s impact on hiring workflows, where machine learning models improve candidate matching beyond keyword searches.

Qingwu does not describe the outcome as something he achieved alone. He credits Jaz with helping him navigate the process, advocating for his experience, and keeping communication moving between the candidate and the company. Effective matching and evaluation also need to assess skills that are not obvious from a standard job description. Better matching can improve candidate quality, especially for hybrid technical and customer-facing roles.

His confidence in the experience is perhaps best demonstrated by what he did afterward. When a friend who was also searching for work asked how Qingwu had found his position, Qingwu introduced him to Jaz. He believed the conversation would be worthwhile for both a capable candidate and a recruiter searching for strong technical talent. 

What This AI Recruiting Success Story Shows Talent Acquisition Employers

Qingwu’s placement offers several broader lessons for companies hiring AI engineers in Canada. It is one of the success stories showing that, when managed well, AI in recruitment can improve speed and candidate quality; 86.1% of recruiters using AI reported faster hiring processes, and companies using AI report 82% better quality hires.

First, job titles do not always reveal the full relevance of a candidate’s experience. Transportation research, automotive engineering, robotics, manufacturing systems, and industrial AI may appear to be separate categories, but they can share foundational technologies such as computer vision, sensor data, deep learning, detection, and tracking. For talent acquisition teams, that makes a well-scoped job description and stronger job descriptions essential to a candidate-focused review of transferable skills.

Second, advanced academic research must be evaluated rather than automatically accepted or dismissed. A PhD is not a substitute for every form of professional experience, but it can involve years of technically rigorous work that should not be categorized as zero experience. The best AI implementation uses objective data to support decision-making, helping HR leaders and HR teams make data driven hiring decisions that improve hiring decisions and reflect company culture.

Third, speed matters. Strong AI candidates may be interviewing with several organizations simultaneously. A recruitment process that stalls between stages can lose a candidate even after the company has identified a good match. In Qingwu’s case, active communication helped move the interviews forward while interest remained high. 

Finally, specialized recruiting requires advocacy grounded in evidence. Jaz could advocate effectively because he had spent time learning what Qingwu had actually done. The recommendation was not based on general enthusiasm. It was based on a clear connection between the candidate’s technical background and the employer’s needs.

Syndesus Finds the AI Candidate Who Might Otherwise Be Missed

The most valuable recruiting outcomes do not always come from finding the person with the most obvious résumé. Sometimes they come from recognizing that a candidate’s experience is more relevant than it first appears.

Qingwu had the technical foundation Maneva needed. Maneva had an opportunity that could turn his research background into practical industrial AI work. The missing element was a partner capable of seeing the connection and ensuring that both sides had the opportunity to evaluate it.

That is the role Syndesus played.

For companies building AI teams in Canada, Syndesus provides access to a curated technical talent network while helping employers assess specialized experience, accelerate interviews, and connect with candidates whose capabilities may be missed by conventional recruiting filters. 

Qingwu’s story demonstrates what can happen when AI recruiting moves beyond résumé forwarding and becomes a process of technical understanding, informed advocacy, and active partnership. Contact us today.

FAQs About This AI Recruiting Success Story