SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: CAD 140,000 - 160,000 / annual
ContactMonkey is hiring an Applied AI Engineer to join the Engineering Team and help build agentic AI capabilities into their employee engagement platform. You'll work alongside senior engineers, Product, and the CPTO to design, prototype, and ship production-grade agentic workflows that improve how internal communications are curated, designed, delivered, measured, and orchestrated.
This is a hands-on, infrastructure-owning role. You won't just call LLMs—you'll help build the platform they run on. The tech stack includes Ruby on Rails and Vue.js in a production SaaS codebase; Amazon Bedrock for model inference; AWS on EKS provisioned with Terraform and Terragrunt; Sidekiq for background work; MySQL and PostgreSQL. The team is mid-migration to GitOps with Argo CD and Karpenter.
Key responsibilities include:
- Design and implement agentic workflows and orchestration patterns (dynamic routing, tool-using agents, feedback loops).
- Own evaluation harnesses for shipped features: build datasets, automated grading, and offline regression suites to measure whether prompt or model changes actually improve outcomes.
- Partner with SRE to build and own AI infrastructure on EKS as code.
- Ensure reliability, cost efficiency, and safety: instrument token spend, latency, and failure rates; design for graceful degradation under hallucination, timeout, rate limit, and cost-blowout scenarios.
- Treat prompts and model configurations as versioned, reviewable, rollback-able artifacts.
- Ensure data residency and trust: keep customer data in-region, log model invocations, and prevent prompt injection and PII exposure.
- Run experiments to benchmark model value and iterate systematically on prompts, retrieval, and configurations based on data.
- Work closely with Product Management to turn ideas into concrete technical work.
- Participate in design discussions, code reviews, and documentation practices.
- Contribute to engineering culture as the team grows.
ContactMonkey powers measurable employee engagement for brands like IKEA, Roku, and KPMG. The all-in-one platform features a drag-and-drop email builder, engagement tools, and analytics. The company has raised $55M in Series A funding and is recognized as one of Canada's fastest-growing companies (Deloitte Technology Fast 50 and Fast 500).
The role is based in downtown Toronto (King and Spadina) with hybrid work 3x per week or as needed. Benefits include 100% employer-paid health benefits, HSA, stock options, RRSP matching, generous vacation, personal development budget, one personal day, two volunteering days, birthday off, and five health days per year.
REQUIREMENTS:
- Bachelor's degree (or higher) in Computer Science, Statistics, Mathematics, or Engineering, or equivalent practical experience.
- 3+ years of professional software engineering experience, including hands-on work with AI/ML or LLM-powered features.
- Strong fullstack experience: Ruby on Rails (or equivalent backend language/framework) and Vue.js/React (or equivalent frontend framework). Comfortable navigating production SaaS codebases.
- Hands-on experience writing and applying infrastructure as code, preferably Terraform (modules, plan review, state management). Terragrunt, CloudFormation, or CDK experience transfers.
- Working knowledge of AWS beyond the console: IAM (roles, policies, least privilege), VPC and networking basics, secrets management, containerized workload deployment and observability. Kubernetes familiarity is a strong plus.
- Demonstrated experience taking LLM work past prototype stage: iterated on prompts against real usage, understood the gap between notebook and production.
- Experience building evaluation beyond trial-and-error: can measure accuracy, safety, and latency; knows when a benchmark win is real versus noise.
- Understanding of LLM call structuring for reliability: function calling, structured outputs, systematic response testing, failure mode design (hallucinations, latency, cost).
- Ability to reason about AI feature costs (tokens, context size, retries, caching) and treat cost as an engineering constraint.
- Familiarity with RAG architectures and vector stores; interest in multi-agent orchestration frameworks.
- Default to using modern AI tools in your own workflow.
- Strong problem-solving approach: question first, then answer.
- Intellectual humility, empathy, and strong listening skills.
- Ability to work collaboratively and independently.
- Comfort with agile development.
- Excellent oral and written communication skills.
Stand-out qualifications:
- Built something agentic (LLM reasoning across multiple steps, tool-using, feedback adaptation; side projects count).
- Built an eval harness that changed a shipping decision.
- Operated AI in production (cost spikes, provider outages, model deprecations, quality regressions under real traffic).
- Bias toward simplicity; pleased when a well-structured baseline beats complexity.
- Experience with Amazon Bedrock or opinions on model routing and provider abstraction.
- Experience in the email ecosystem or similar high-scale communication platforms.
- Built with AI coding agents and have views on their strengths and weaknesses.
- Prototype fast, iterate quickly, use data to change direction.
- Share learning with peers via writeups, demos, docs, or internal talks.
