Senior Full Stack Agentic Developer, Contract

BIMM • Toronto, ON

1d • CA$60 - CA$80 • Hybrid

About The Position

BIMM is a Toronto-based agency that lives at the intersection of data, technology, and creativity to provide clients with high-performing campaigns and digital products that shape connected customer experiences. We’ve helped our clients implement fully integrated CRM customer journeys, built robust digital experiences, and created national loyalty programs with decades of experience in sectors of automotive, financial, telco and retail. We are creative realists, which means our ideas and solutions are grounded in the realities of our clients’ business. We believe big ideas are nothing without big results (and flawless execution); high-performing agencies deliver high-performing campaigns, products, and experiences; and the best solutions happen when we collaborate with clients as partners. Our Technology department is a 50+ person distributed team of Developers, QA, and Architects delivering amazing digital products on a leading-edge technology stack. This is an pipelining opportunity - we are anticipating our team growing.

Requirements

• Be available to collaborate with the team in our Toronto office 2 days per week.
• 5+ years of working knowledge with modern frameworks and languages (REACT & NodeJS)
• Familiarity with OAuth2/OIDC, API key management, and access audit trail patterns
• Experience with SSR applications
• Experience with GraphQL and Apollo
• Experience building or integrating MCP-compatible REST/GraphQL APIs; familiarity with tool interface contracts, versioning, and consumer documentation.
• Familiarity with SLO definition and basic observability practices (logs, metrics, distributed traces) for production API services.
• Hands-on experience with LLM structured output patterns: function/tool calling
• Familiarity with LLM evaluation approaches: building golden test sets, LLM-as-judge pipelines, and prompt regression testing; ability to quantify output quality with measurable metrics rather than vibes.
• Experience building agentic workflows — multi-step tool chains, state machine-based agents, or orchestration frameworks (LangGraph, LangChain, AutoGen, or custom) — with an emphasis on deterministic routing and graceful failure handling

Nice To Haves

• Demonstrated experience defining and rolling out engineering standards at a team or org level (coding conventions, PR workflows, testing mandates, API contracts)
• Experience managing and growing engineers — performance conversations, career development, structured feedback
• Track record of driving AI tool adoption within an engineering team — not just using the tools but creating the onboarding, guidance, and culture around them
• Comfort operating at the intersection of technical leadership and delivery management — you can write the ADR and run the retro
• Familiarity with AWS & Docker
• Experience with Styled Components
• Experience with prompt caching, semantic routing, or output memorization strategies to reduce non-determinism at scale.
• Ability to instrument LLM calls with structured traces (input, output, latency, token count, tool calls invoked) using frameworks like LangSmith, OpenTelemetry, or custom logging; can define SLOs for agent task success rates

Responsibilities

• Optimizing website performance: Making sure websites load quickly and efficiently.
• Developing APIs (Application Programming Interfaces): Creating interfaces that allows different applications to communicate with each other.
• Understanding the entire web development process: Having a holistic view of how the front end and back end work together.
• Working with different technologies: Being proficient in a variety of programming languages, frameworks, and tools.
• Build and maintain automated evaluation pipelines (evals) for agent and skill outputs — including LLM-as-judge scoring, regression test suites, and golden dataset validation — so that prompt and model changes are measurable before they ship.
• Define strict input/output contracts for MCP tools and agent skills using typed schemas (TypeScript interfaces, JSON Schema); ensure tools handle edge cases, surface structured errors, and never return ambiguous output that an LLM must interpret.
• Own the prompt engineering lifecycle for assigned tools and skills — version-controlled prompt templates, parametric input injection, and structured system/user role separation — ensuring prompts are testable, reproducible, and free of implicit context drift.
• Contribute to the team's MCP tooling catalog — implement, test, and document MCP-compatible API integrations (particularly GraphQL/Apollo and developer portal tooling); participate in rollout readiness reviews including SLO definition and support contact documentation.

Benefits

• Our squad is fun, friendly, and entirely egoless. Our social committee plans BIMM bashes, Tasty Thursdays and quarterly outings like candle making workshops, board game nights, Jays games and cooking classes. Our summer and holiday parties are quite memorable as well! Having Fun Everyday is one of our core values.
• Our dedicated DEI committee provides thought -provoking insights which are reflected in our work; They also organize engaging awareness activities and events to showcase BIMMER’s professional talents and personalities. What talents might you bring? 😊
• BIMM is part of the Kyu Collective which gives our employees access to the network’s resources, training, offices, and more. We also prioritize your personal and professional development with opportunities like on-site Lunch n’ Learns, conferences and online courses. Love 2 Learn.
• BIMM is committed to providing an environment that is inclusive and accessible. We are an equal opportunity employer and consider all applicants for employment without discrimination. Please let us know if accommodation for the recruitment/interview process is required and we will work with you to make sure your needs are met.
• During the challenge portion of our process, our team uses AI to analyze your repo and provide feedback. Agent outputs are always reviewed by our teams to ensure accuracy. Candidates are then screened by humans, and approved by humans.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed