Make eeze smarter and safer: retrieval, grounding and guardrails for an AI that negotiates with real money on the line.eeze is an AI checkout. It answers buyer questions from a company's own docs, handles objections and can negotiate, but only inside rules the founder wrote, with every offer validated server-side. That combination of useful and safe is the whole product, and it is an engineering problem more than a prompt problem.You will own the AI layer: retrieval quality, grounding and citations, evaluation, guardrails and cost. You will also help clients ship AI features in their own products during consulting engagements.What you will doImprove retrieval and grounding so answers cite the tenant's knowledge base accuratelyDesign and enforce guardrails around offers, discounts and claimsBuild evaluation suites that catch regressions before customers see themTune latency and cost across Anthropic and OpenAI models, including bring-your-own-key setupsTurn knowledge-gap analytics into a product feature founders act onPrototype and ship AI features with clients during engagementsStay current on model capabilities and separate the helpful from the noiseWhat we are looking forStrong software engineering fundamentals; this is a production engineering role,
not a research roleHands-on experience shipping LLM features to real usersPractical knowledge of retrieval, embeddings, context management and evaluationExperience with at least one of the major model APIs in productionA healthy scepticism and the habit of measuring model behaviour instead of assuming itAble to work from our Toronto office part of the weekNice to haveExperience with structured outputs, tool use and multi-step agent flowsBackground in payments, commerce or other domains where mistakes cost moneyFamiliarity with guardrail and validation patterns for user-facing AIExperience running evals in CIWho you areYou treat the model as a component, not a magic boxYou would rather ship a reliable small feature than demo an impressive fragile oneYou test with adversarial inputs because customers willYou write down what you learn so the team compoundsYou care whether the answer is actually trueHow we hire1 Intro call with an engineer, about 30 minutes2 Short take-home assignment around retrieval and grounding on real docs3 Technical conversation about your assignment and AI features you have shipped4 Conversation with the founders5 Offer
📌 Ai Engineer (Toronto)
🏢 CLR3
📍 Toronto