AI Engineer

Title: AI EngineerSalary: Based on experience + Benefits + VacationLocation: Thornhill, Ontario, Canada (Remote) Length: Permanent, full-timeAbout Us:At WellnessLiving, we empower thousands of health and wellness business owners to turn their entrepreneurial dreams into reality. Our mission-critical software fuels their vision, supporting millions of clients around the world in their wellness journeys. With a deep commitment to putting our customers first, we foster a culture that values high performance, adaptability, and accountability. If you are a skilled professional who thrives in a fast-paced, customer-focused environment and are passionate about making a meaningful impact on the health and wellness industry, we would love to connect with you.     Our team is driven by four core values that shape everything we do. If you share these values and meet the qualifications outlined for this role, we encourage you to apply - we'd love to learn more about you    Customer First – We approach every challenge with a customer-focused lens, driven by an obsession with our customers' happiness and success.    Excellence – We approach every task, whether big or small, with a steadfast commitment to exceptional execution and the pursuit of greatness.    Accountability – We take full ownership of our decisions, actions, and outcomes – both successes and failures.    Adaptability – We recognize that sustained success demands that we be malleable and purposefully evolve, acknowledging that the world is dynamic and constantly changing. About You: As an AI Engineer, you'll build production AI features across four modules—Front Desk Agent, Business Coach, Sales & Marketing Suite, and Omnichannel Message Center. A core part of the role is prompt engineering: designing reusable prompt templates and tool schemas, grounding with business data, and running automated evaluations to deliver low-latency, observable microservices connected to real APIs (booking, calendars, payments, messaging, CRM).Key Responsibilities:Design & ship AI features across modules; translate user journeys into secure, low-latency microservices backed by an LLM.Prompt engineering: develop prompt templates (zero/one/few-shot), function/tool calling schemas (JSON), and guardrails/policies for bookings, FAQs, and safe human handoff; version prompts and track telemetry (win rate, latency, cost).Conduct experiments to evaluate model performance, benchmark across Gemini, GPT, Claude, LLaMA, and other LLMs, and recommend best-fit solutions. Grounding & retrieval: bind conversations to client profiles, calendars, memberships, pricing, and staff availability; implement context windows, citations, and fallback strategies.Voice & telephony: integrate STT/TTS and Twilio/LiveKit with barge-in, interruption handling, and dialogue-state recovery.Booking & payments: connect to the WellnessLiving engine, calendar sync, conflict/resourcing logic, and Nuvei/Paragon flows with retries and failure recovery.Insights & growth: ship Business Coach KPIs and NL Q&A; add predictive/"what-if" scenarios and personalized recommendations.Marketing automation: segmentation, AI content, A/B tests, send-time optimization, lead scoring; close the loop with engagement signals to tune prompts/models.Omnichannel inbox: unify SMS/email/in-app/WhatsApp/voice/social with intent/sentiment routing and in-thread booking/payment; meet audit/SLA targets.Engineering excellence: write tested Python/TypeScript; Dockerize; add logs/metrics/traces/alerts; maintain runbooks/postmortems and participate in on-call.Evaluate & harden: build golden-set tests, automated prompt/model evals, red-team for jailbreak/injection, and optimize accuracy/latency/cost/resilience.Minimum Qualifications:Bachelor's or Master's degree in Computer Science, Data Science, AI, or related field. 3+ years building backend or ML-enabled apps with production LLM exposure.Proficiency in Python and one of ; REST/GraphQL experience.Demonstrated prompt-engineering experience, tool/function calling, structured outputs (JSON), retrieval/RAG prompting, safety/guardrail design, and A/B or golden-set evaluation.Proficiency in Python and frameworks such as PyTorch, TensorFlow, Hugging Face Transformers, LangChain, or LlamaIndex. Strong experience with data manipulation and analysis tools (SQL, Pandas, NumPy). Familiar with Docker, Git, CI/CD, cloud services; comfortable adding observability and tests.Solid authN/authZ, privacy, and secure PII handling; benchmarking for latency/concurrency/error recovery.Nice to Have:STT/TTS, telephony (Twilio, SIP) or WebRTC (LiveKit).RAG pipelines, latency-aware caching, and prompt evaluation frameworks (e.g., Ragas, OpenAI Evals) or libraries (Guidance, DSPy, Semantic Kernel, LangChain).AWS (Lambda, S3, Aurora Serverless, Redshift), Kubernetes; analytics/KPI modeling.CRM/marketing (Salesforce/HubSpot/Zapier; Google & Meta Ads); messaging (Twilio, SendGrid, FCM/SES); basic React for AI-assisted UI.How We'll Measure Success:Higher prompt win rate on core tasks; reduced hallucinations and safety incidents.Front Desk Agent: autonomous booking with fast responses.Business Coach: strong intent recognition; real-time insights that improve KPIs.Marketing Suite: accurate lead scoring/personalization; low-latency triggers with clear ROI.Message Center: accurate triage/routing;