Mitacs Accelerate 2026: The Complete Guide to Canada's Research-Industry Internship

Mitacs Accelerate is the single highest-leverage research-funding play available to a Canadian SME that has a research-flavoured project and a willingness to work with a university. The mechanics are simple enough to sketch on a napkin: you find a graduate student or postdoc at a Canadian post-secondary institution whose research interests align with a problem you want solved, your company contributes $7,500 per 4-month unit, Mitacs matches with another $7,500, and the intern receives a $15,000 stipend to spend roughly half their time embedded with your team and the other half back in their lab applying academic methods. Units stack — six months, twelve months, twenty-four months — and for the company-paid portion of the stipend that supports SR&ED-eligible work, you can layer the federal investment tax credit on top. This guide walks through how Accelerate actually works in 2026, the 2024 partner-contribution change that quietly halved the cost of entry, the application flow, and where the program interacts with SR&ED.

What Mitacs Accelerate actually is

Mitacs is a federally and provincially funded non-profit whose mandate is to connect Canadian industry with the country's academic research capacity. Accelerate is its flagship program and has been the entry-point for tens of thousands of industry-academic research collaborations in Canada. The structural idea is older than most of its current participants: rather than asking companies to fund full graduate scholarships outright, or asking universities to find industrial relevance for their PhD students retroactively, Accelerate splits a defined research project into 4-month "units" co-funded between the company and Mitacs and supervised jointly by the company and a faculty member.

Three details make the program genuinely distinctive from the broader landscape of Canadian research funding:

• The intern is the deliverable. Unlike NSERC Alliance or IRAP, where the deliverable is a research output and the personnel are inputs, in Accelerate the intern's training experience is itself part of the program's purpose. This shifts the review criteria away from pure commercial novelty and toward research quality and trainee value.
• The company pays the intern, not the university. The $15,000 stipend goes to the intern. Your $7,500 contribution and Mitacs's $7,500 match are pooled and disbursed as a stipend, typically through the academic institution. There are no project overheads, no equipment budgets, no PI salary lines — just an intern's time, paid.
• Rolling intake. There is no fixed deadline. You apply when your project is ready, and Mitacs typically returns a decision in 6–8 weeks. For SMEs with project timelines that don't align to annual federal grant cycles, this is the program's largest practical advantage.

The $7,500 partner contribution — and the 2024 change

Until mid-2024, the standard minimum partner contribution for an Accelerate unit was $15,000 per 4-month internship, with Mitacs matching at the same level for a total $30,000 stipend. In 2024, Mitacs reduced the minimum partner contribution to $7,500 per unit, with a corresponding adjustment to the Mitacs match at $7,500 — total stipend $15,000.

This is the most consequential change to the program's economics in years. It cut the cost of entry in half. For a small company evaluating a one-unit project, the calculation went from "$15,000 outlay for $30,000 of research work" to "$7,500 outlay for $15,000 of research work" — the leverage ratio is identical but the absolute cash exposure is far easier to absorb. For SMEs with annual revenues under $5M, this puts the program inside the discretionary R&D budget of companies that would have hesitated at the old level.

A few practical notes about the contribution mechanics. Mitacs accepts higher partner contributions if the company wants to fund a richer stipend (some companies top up the intern to align with internal pay bands, particularly for PhD-level work in software or biotech). The matching ratio doesn't scale linearly above the minimum — Mitacs's contribution is a fixed match, not a percentage match. The minimum is the leverage point; above it, additional company dollars flow through directly to the intern.

How to find a university research partner

The single most common reason an Accelerate application stalls at the starting line is that the company has the project idea but no academic partner. The program structurally requires both: you cannot apply alone. You need a faculty member at a Canadian post-secondary institution who is willing to supervise the intern and whose research expertise aligns with the work.

For most SMEs, the practical paths to finding a partner are:

• Mitacs business development team. Mitacs employs a regional business development team whose explicit job is to help companies identify suitable academic supervisors. This is the lowest-friction starting point. You describe the research problem; they connect you with faculty in their network whose work touches the problem area.
• University industry liaison offices. Every research-intensive Canadian university has an industry liaison or research partnerships office. They can match you with faculty across departments. Larger institutions (Toronto, Waterloo, McGill, UBC, McMaster) tend to be the most responsive; smaller universities are often more flexible on project scope.
• Direct outreach. If you know your problem area well, faculty pages and recent publication lists let you identify researchers working on adjacent problems. A specific, technically credible cold email to a faculty member — one that demonstrates you've read their work and articulates a defined project — gets responses at rates that surprise people who haven't tried it.
• Existing graduate or postdoc relationships. If a current or former employee is finishing a graduate degree, or a co-op intern is returning to grad studies, those existing relationships can convert into an Accelerate unit with the trainee's academic supervisor.

Eligibility — Canadian SMEs and non-profits

The "research component" requirement is worth elaborating. Mitacs is not a wage subsidy program — it is a research program. The project must have a question or hypothesis that the intern can investigate using research methods. "Build us a website" doesn't qualify; "evaluate which architectural patterns optimize for X" can. The line is usually clearer than companies expect: if a faculty member would be willing to supervise a graduate student on the problem, it's a research problem.

On the trainee side, the intern must be a graduate student or postdoctoral fellow at a Canadian post-secondary institution. Undergraduate students are not eligible for Accelerate (Mitacs's Globalink program serves a separate undergraduate audience). The intern's academic supervisor must be eligible to hold research grants at their institution.

The application flow and the ~6–8 week review

Accelerate applications run through the Mitacs portal. The application is a joint submission — the company, the academic supervisor, and the intern all contribute sections and all must approve the final submission. Once submitted, the application enters peer review.

• Step 1 Identify the intern and supervisor. The intern is named in the application — this is not a program where you apply first and recruit later. By submission time you need the specific graduate student or postdoc, their supervisor, and their host institution.
• Step 2 Scope the research project. Draft a project plan with deliverables, methodology, and a clear research question. Define how many units (one 4-month unit, or stacked units for longer engagements) and the total budget.
• Step 3 Submit through the Mitacs portal. The company, supervisor, and intern each complete their respective sections. The supervisor's institution approves and sponsors the submission.
• Step 4 Peer review. Mitacs sends the application to external academic reviewers for assessment of research quality, training value, and supervisor capacity. This is the part of the timeline that varies most — straightforward applications move quickly, applications with weak research framing or supervisor-fit questions take longer.
• Step 5 Decision in ~6–8 weeks. Typical end-to-end review is six to eight weeks. Outcomes: approved, approved with conditions, revise-and-resubmit, or declined. The revise-and-resubmit pathway is common and not penalizing — reviewers will tell you what to strengthen.
• Step 6 Internship begins. Once approved, the company pays its contribution, Mitacs disburses the match, and the intern begins the 4-month unit on the agreed start date.

A few practical observations about timing. Companies that have run Accelerate before tend to underestimate how long the up-front scoping takes for a first project — not because Mitacs's process is slow, but because aligning the company's commercial timeline with the intern's academic calendar and the supervisor's grant cycle is a coordination problem. If the project needs to start September 1, work backwards from there: an application submitted in early summer is the comfortable timing; one submitted in late August is rushed.

Stacking units for longer projects

The 4-month unit is the program's atomic structure, but most meaningful industry research takes longer than four months. Accelerate addresses this through stacking: multiple units strung together into a single longer engagement, with a single application or multiple sequential applications.

• Two units — an 8-month engagement at a $30,000 stipend ($15,000 partner + $15,000 Mitacs match). Common for projects with a defined research question and a build-and-test cycle.
• Three units — a 12-month engagement at $45,000 ($22,500 partner + $22,500 Mitacs match). Typical for a year-long applied research project that aligns with a graduate student's thesis chapter.
• Six units — a 24-month engagement at $90,000 total stipend ($45,000 partner + $45,000 Mitacs match). The structure that supports a PhD candidate's industrially-relevant thesis work or a postdoc's two-year industry-embedded research project.

Stacking is not automatic — the longer the project, the more carefully Mitacs scrutinizes the research narrative for whether the work genuinely requires the extended duration. A four-unit project framed as "more time to do the same thing" is weaker than a four-unit project framed as a phased research program where each unit answers a distinct question that feeds the next.

Grad students vs postdocs — choosing the trainee level

Accelerate accepts both master's students, PhD candidates, and postdoctoral fellows as interns. The choice between them is more strategic than it first appears.

• Master's students are typically in the program for 18–24 months total. Their research training is still developing, but they're highly motivated and bring fresh exposure to current methods. Best fit: defined, scoped research problems where a clear deliverable in 4–8 months is the goal.
• PhD candidates bring deeper research training and a longer time horizon. They're already steeped in the literature and methodology of their field. Best fit: longer engagements (multiple stacked units) where the work can become a thesis chapter, or projects that demand sophisticated research methodology.
• Postdocs are the most experienced trainee category — PhD-trained researchers with established publication records. The cost structure remains the same ($15,000 stipend per unit), which makes postdoc engagement extraordinarily cost-effective relative to hiring them directly. Best fit: high-stakes research problems where deep technical expertise is the bottleneck, or projects that could lead to recruiting the postdoc into a permanent industry role.

The hiring-conversion angle is worth flagging explicitly. A non-trivial fraction of Accelerate engagements end with the company hiring the intern into a full-time role after the unit concludes. For SMEs in deep-tech areas where senior research talent is scarce and expensive, treating Accelerate as a paid evaluation period — both the company evaluating the trainee and the trainee evaluating the company — is a strategy that quietly drives a large share of the program's economic value.

Stacking Mitacs with SR&ED

This is where Accelerate becomes genuinely high-leverage for a company doing R&D in Canada. The Scientific Research and Experimental Development (SR&ED) tax credit can apply to the company-paid portion of the Accelerate stipend, provided the work the intern performs while embedded at your premises is SR&ED-eligible under the CRA's IC 2012-02 framework.

The mechanics:

• The Mitacs match is government assistance. The $7,500 contributed by Mitacs is non-dilutive funding, but for SR&ED purposes it is "government assistance" and must be deducted from any SR&ED-eligible expenditure pool that the same dollars fund. You cannot claim SR&ED on Mitacs's $7,500.
• The company's $7,500 may be eligible. The portion of the stipend funded by your company — the $7,500 per unit — supports the intern's time. Where that time was spent on SR&ED-eligible activity, conducted in Canada, the expenditure is potentially eligible for SR&ED.
• Intern status matters. The intern is typically classified as a contractor or fee-for-service researcher, not a T4 employee of your company. SR&ED treatment of contract expenditure follows different rules than salary — specifically, only 80% of arm's-length contract amounts qualify, and non-arm's-length contracts have their own rules. Get this classification right at the start of the engagement.
• Documentation has to match. Time the intern spent at your premises on SR&ED work is the claimable activity. Time spent at the university lab on academic literature review, thesis writing, or training that is general to the intern's program of study is not claimable. The intern needs to keep contemporaneous time records that distinguish between the two.

The S/THERI framework — Scientific or Technological uncertainty, Hypothesis, Experimentation, Results, Iteration — is the structural test the CRA expects to see in a defensible technical narrative. Mitacs research projects already lend themselves to this structure better than most internally-conducted R&D, because the academic supervisor's involvement essentially forces the work to articulate a research question, a methodological approach, and an evidence-based evaluation. If you scope the Accelerate project deliberately around an uncertainty your engineering team has been unable to resolve through standard practice, the same research narrative becomes the spine of your SR&ED claim for the year. Two birds, one carefully-constructed research question.

One additional consideration worth flagging: the academic supervisor and the host institution generally cannot themselves be the SR&ED claimant for the work — the SR&ED claim is yours, the company's. The intern's contractor agreement, the project workplan, and the company's time records all need to be in your hands and aligned to your tax year. If the academic institution maintains the contractor relationship with the intern (which is the typical Mitacs structure), make sure your engagement documentation captures the contracted research services rendered to your company specifically — not just the broader Mitacs arrangement.

Common reasons applications stall

From the pattern of Accelerate applications that don't get approved on first submission, the recurring failure modes are predictable:

• The project is consulting, not research. If the work doesn't have a research question that an academic reviewer would recognize as a research question, the application fails the threshold test. "Help us solve X" isn't enough; "investigate whether approach A or approach B better resolves the constraint X under condition Y" is closer to the bar.
• Weak supervisor-project fit. The faculty member is willing but their research expertise doesn't actually align with the problem. Reviewers can tell. Pick a supervisor whose published work overlaps the problem area, not just whose general field is adjacent.
• The intern is doing work they couldn't already do. Mitacs is a training program. If the intern is being asked to do work they've already mastered — a senior PhD doing the same coding work they did in their previous role — the training value is weak. The work should stretch the trainee.
• Insufficient industry research narrative. The company section reads like marketing copy rather than a research scope. Reviewers want to see that the company has thought concretely about the research question, the methodology, and the deliverables — not just stated that they need help.
• Mismatched timeline. The proposed start date is too soon for Mitacs's typical 6–8 week review window. This is rarely a fatal flaw but causes friction.
• Inadequate intellectual property arrangement. Mitacs requires a clear IP arrangement between the company, the university, and the intern. If this hasn't been worked out, the application is held until it is. Start the IP conversation with the university's industry liaison office early.

One pattern that's worth calling out separately: companies sometimes submit Accelerate applications for work they've already started. Mitacs is explicit that the funded work must be prospective — the intern's time on the project cannot begin until the application is approved and the unit officially starts. Backfilling a Mitacs application around work already underway puts the application at risk and, more importantly, puts the SR&ED claim for that activity at risk too if the funding structure is later challenged. Apply before the work begins.

A related pattern: companies that have used Accelerate successfully before sometimes get pattern-locked into one academic supervisor and one host institution. The supervisor was great for the first project, the relationship is comfortable, the next project goes to the same supervisor regardless of fit. This works fine when the new project is in the supervisor's wheelhouse; it works poorly when it isn't. Treat each new Accelerate engagement as a fresh decision about supervisor-project fit, not a continuation of an existing relationship.

Related Mitacs programs — Accelerate Entrepreneur and BSI

Accelerate is the flagship, but Mitacs runs a small family of related programs that fit specific use cases. Two are worth knowing about for SMEs:

• Mitacs Accelerate Entrepreneur is an Accelerate variant designed for graduate students and postdocs whose research is becoming a startup. The intern is also a founder; the partner organization is their own early-stage company. The structure mirrors Accelerate (4-month units, matched funding) but accommodates the founder-researcher dual role that doesn't fit a traditional industry-academic partnership. For research-led founders coming out of Canadian universities, this is the path of least resistance into the Mitacs system.
• Mitacs Business Strategy Internship (BSI) is a separate program oriented toward MBA students, business school graduates, and trainees in business-adjacent disciplines. The internship focuses on business strategy, market research, commercialization planning, and similar applied business problems — areas that don't fit cleanly into Accelerate's research-focused mandate. For SMEs that need strategic project work rather than scientific research, BSI is the right Mitacs program to look at.

Mitacs also runs Globalink (international undergraduate research internships in Canada), Elevate (a longer-form postdoctoral fellowship program), and various training programs in entrepreneurship, business strategy, and research skills. The full program catalogue is on the Mitacs website. For most Canadian SMEs running R&D projects with a research component, Accelerate is the program to start with.

Final thoughts

Mitacs Accelerate is one of the most cost-effective ways for a Canadian SME to bring research talent into the company without committing to a full hire and without navigating a complex federal grant application. The 2024 reduction in the partner contribution to $7,500 per unit changed the economics for smaller companies in particular — the program now sits comfortably within the discretionary R&D budget of companies that would have stretched for the old $15,000 entry point.

The program's structural design rewards three things: a clearly-articulated research question, a well-matched academic supervisor, and a trainee whose work the company is prepared to invest in beyond the unit itself (whether through hiring, follow-on units, or research collaboration). The companies that get the most value out of Accelerate aren't the ones treating it as a wage subsidy — they're the ones treating it as a low-risk way to access research methods, talent pipelines, and university research infrastructure that would otherwise be hard to reach.

Stacking Accelerate with SR&ED on the company-paid portion of eligible work is where the leverage compounds. A well-designed multi-unit Accelerate engagement embedded in an SR&ED-eligible research program can generate a return that's substantially better than either funding source alone — provided the documentation and the IC 2012-02 framing are sound.

Mitacs Accelerate is most powerful when paired with other Canadian R&D supports — SR&ED, IRAP, NSERC Alliance, provincial innovation programs. Use the Grant Finder to see the full stack of Canadian research funding available to your team.