AI Architect | Lead, Agentic Systems

AI Architect | Lead, Agentic Systems & GenAI

Canada; US

We're hiring a AI Architect to design and deliver agentic, LLM-powered systems for our clients. This is a hands-on, client-facing role focused primarily on generative AI and agentic workflows, with traditional machine learning as a bonus, not the core.

Role Overview

You will own the end-to-end technical design and implementation of GenAI solutions: from discovery and use-case shaping, through architecture and prototyping, to product ionization and reusable patterns for our consulting teams. The emphasis is on agentic systems, tool-using agents, and workflow automation across our clients' existing platforms and data.

Key Responsibilities

• Lead technical discovery with clients to identify high-value GenAI and agent use cases tied to concrete business outcomes.
• Translate fuzzy ideas into clear solution designs, user journeys, and MVP scopes that can be quickly validated.
• Design end-to-end architectures for GenAI applications: frontend, backend/APIs, orchestration, LLM providers, vector databases, and integrations with enterprise systems and SaaS platforms.
• Build and maintain LLM-powered services: conversational copilots, workflow agents, embedded assistants, and task-specific bots for internal and external users.
• Design, implement, and operate agentic systems: planner/executor patterns, tool-using agents, and (where appropriate) multi-agent patterns for complex workflows.
• Integrate agents with real tools and systems (REST/GraphQL APIs, internal microservices, workflow engines, data platforms), including authentication, authorization, and auditing.
• Establish standards for prompt and system design, tool schemas, safety guardrails, observability, and reliability for GenAI and agentic solutions across projects.
• Mentor engineers and consultants, review designs and code, and drive best practices and shared patterns across multiple client engagements.
• Create reusable reference architectures, templates, and frameworks that accelerate future GenAI and agent projects.
• Contribute to thought leadership via internal enablement and external content (talks, blog posts, OSS) when appropriate.

Must-Have Qualifications

Experience and Background

• 6–10+ years of professional experience as a software engineer, backend engineer, or solutions/enterprise architect.
• Proven track record shipping production-grade backend systems and APIs (not just prototypes or research notebooks).
• Strong programming skills in at least one major backend language (e.g., Python, TypeScript/Node, Java/Scala), with solid engineering practices (testing, code review, CI/CD, version control).
• Demonstrated experience with agentic engineering practices — i.e., AI-native development workflows such as using LLM-powered coding assistants, AI-driven code generation, and prompt-driven prototyping as core parts of the software development lifecycle.
• Significant experience with at least one major cloud provider (AWS, Azure, or GCP), including designing and operating services using containers and/or serverless, logging, metrics, and alerting.

GenAI / LLM Expertise

• Hands‐on experience building applications on top of hosted LLMs (e.g., OpenAI, Azure OpenAI, Anthropic, AWS Bedrock, Gemini, or open‐source models via hosted platforms).
• Strong prompt and system message design skills for chats, copilots, and task automation, including iterative refinement and evaluation.
• Familiarity with embeddings and vector databases (e.g., Pinecone, Weaviate, pgvector, Redis, OpenSearch) and retrieval‐augmented generation (RAG) patterns: chunking strategies, metadata, and relevance evaluation.
• Understanding of GenAI‐specific evaluation concerns: hallucinations, safety controls, relevance, and UX patterns for user control and correction.

Agentic Systems and Tools

• Prior experience building agentic systems, including:
• Planner/executor patterns and multi‐step reasoning flows.
• Tool‐using agents that call external APIs, services, and workflows.
• (Optional but valued) Multi‐agent setups such as supervisor/worker or specialist agents.
• Practical experience with at least one agent/orchestration framework or pattern (e.g., LangGraph, LangChain agents, Semantic Kernel, custom orchestrators, or major LLM providers' tool/agent APIs), or workflow automation platforms with AI/agent capabilities (e.g., n8n).
• Ability to design robust tools: clear schemas, input/output contracts, validation, rate‐limiting, and guardrails for safe execution.
• Strong focus on reliability in agent workflows: idempotency, retries, fallbacks, circuit breakers, timeouts, and safe failure modes.
• Experience implementing observability for agents: logging of tool calls and reasoning trac