The headlines have gone to the hyperscalers. Microsoft, Google, and a growing queue of colocation operators have been snapping up Fraser Valley land and power agreements at a pace that has reshaped the region's industrial real estate market. However, the more durable business opportunity may be quieter, and closer to home.

A distinct cohort of Vancouver-based startups is building the software infrastructure layer that those facilities require: workload orchestration, GPU cluster management, inference optimization, and energy monitoring tools. Several have closed seed rounds in Q1 2026 largely under the radar, drawing capital from both Canadian and U.S. institutional pre-seed funds. The cluster is early, but the structural conditions for something durable are in place.

Why here, why now

The timing is not accidental. The Fraser Valley's planned data centre capacity now exceeds an estimated 800MW across announced projects—a buildout creating an unusually concentrated base of potential enterprise customers within a two-hour drive of Vancouver's startup core.

That proximity matters. Infrastructure software is not sold from a pitch deck; it is sold through repeated technical conversations with operators solving real problems, such as cooling inefficiencies, GPU utilization gaps, and energy cost spikes during inference workloads. Founders who can drive to Abbotsford or Langley to sit with a data centre operator's engineering team possess a customer-discovery advantage that founders in Toronto or Montréal do not. The feedback loop is faster, and the product iteration is tighter.

The global market for GPU cluster management software is projected to grow at a compound annual rate of roughly 28% through 2028, reflecting the surge in inference workloads as enterprise AI deployments move from pilot to production. That tailwind is real, but so is the competition. U.S.-based players with deeper pockets are building in the same space. The window for Vancouver founders to establish customer relationships and switching costs before the market consolidates is measured in months, not years.

The companies to watch

Specific seed-stage companies in this cohort are being tracked through the BC Tech Association's membership directory and Innovate BC's investment tracking, with deal flow surfacing through PitchBook's Vancouver AI infrastructure deal data. These names are not yet household—which is the point. This is the pre-brand phase of what could become a recognizable cluster.

The common thread across the cohort is a focus on the operational layer rather than the model layer. These companies are not building foundation models or fine-tuning pipelines. They are building tools that make the underlying compute run efficiently: scheduling software that minimizes GPU idle time, monitoring dashboards that flag thermal anomalies before they become outages, and energy optimization engines that help operators manage power costs under variable grid pricing.

Several have structured their early contracts as usage-based SaaS agreements with enterprise data centre operators, a model that creates recurring revenue with high switching costs once the software is embedded in operational workflows. That is the kind of unit economics profile institutional investors seek in a B2B infrastructure play.

The BC Centre for Innovation and Clean Energy has been active in compute infrastructure partnerships, providing an additional validation pathway for startups whose tools touch energy management—a category increasingly relevant as data centre power consumption draws regulatory attention.

The capital gap

A structural problem remains. Ontario's $200-million Next Ontario Fund provides that province with a dedicated vehicle for backing technology infrastructure companies at scale. BC has no equivalent. The provincial government has not yet created a dedicated AI infrastructure fund, meaning Vancouver founders in this space compete for capital against Ontario peers who have access to a purpose-built provincial backstop.

The gap has not stopped seed rounds from closing—U.S. pre-seed funds have filled some of the slack, and New Ventures BC's competition cohorts have included infrastructure software entrants. However, the absence of a provincial fund creates a Series A cliff. Seed capital secures product-market fit; growth capital secures enterprise contracts. Without a provincial anchor, BC founders may find themselves shopping those rounds south of the border—and potentially relocating with them.

The bigger picture

Infrastructure software is historically where the most durable B2B SaaS companies are built. The switching costs are real, the contracts are long, and the revenue is recurring. The customers—data centre operators managing hundreds of megawatts of compute—are not going anywhere.

Vancouver has the talent base, the proximity to a rapidly expanding customer set, and a nascent cohort of founders building in this space. What it needs is a clearer institutional narrative from investors, government, and the ecosystem that this cluster is worth backing at scale before U.S. competition arrives in force.

The hyperscaler land rush got the headlines. The software layer is where the equity is built.