AI video intelligence platform for retail loss prevention and operations
Solink processes video + operational data through AI models (PyTorch, TensorFlow) to detect fraud, theft, and inefficiencies in real time. The tech stack reveals a mature, distributed system: Kubernetes orchestration, serverless (Lambda, SQS), managed databases (DynamoDB, PostgreSQL, Snowflake), and enterprise integrations (Salesforce, NetSuite, ADP). Sales hiring (6 roles) outpaces engineering (5), reflecting a sales-led go-to-market focused on expanding restaurant and cannabis retail verticals, with concurrent work on agent orchestration and LLM/VLM integration suggesting internal AI capability-building.
Solink is an AI-driven video intelligence platform founded in 2009 and based in Ottawa. The product connects video streams, POS systems, and operational data to automatically detect losses, safety risks, and operational gaps across multi-location retail environments. The platform serves security, loss prevention, and operations teams at mid-market retailers, with active expansion into restaurant and licensed cannabis verticals. The company operates a 201–500 person team across engineering, sales, finance, and research functions, with hiring accelerating across Canada, the UK, and the US.
Solink uses TypeScript, React, Node.js, Python, and Rust for application development; PyTorch and TensorFlow for AI models; AWS (EKS, Lambda, SQS, Kinesis), GCP, and Azure for cloud infrastructure; PostgreSQL, DynamoDB, and Snowflake for data; and Salesforce, NetSuite, and ADP Workforce Now for enterprise operations.
Solink is actively growing in restaurants and licensed cannabis retail. The company is developing strategic account plans and sales motions for these segments alongside core loss prevention and operations use cases.
Other companies in the same industry, closest in size
Solink's technology stack, projects, and hiring signals are inferred from public hiring and company data — career pages, public listings, and company web presence — then clustered and de-duplicated. Figures are estimates that refresh over time. Read our full methodology →
This is not an official vendor or customer list. It is a technology-adoption signal inferred from public data, intended for B2B research.