Intermediate AI Engineer

I am hiring an Intermediate AI Engineer on a 6–12-month contract on behalf of a well-established organization in Toronto.

In this role, you will help build and scale company’s next-generation Agentic AI chatbots and systems while contributing to a growing portfolio of GenAI applications designed to solve real-world problems. You will bridge the gap between ambiguous LLM capabilities and deterministic, high-performance production software, will be designing autonomous agent workflows using Google ADK, engineering dynamic context pipelines, and deploying robust backend services.

Key Responsibilities:

Agentic AI & GenAI Development

• Agent Orchestration: Design and implement intelligent agent workflows (e.g., tool-calling, planning, multi-agent collaboration, and self-reflection) primarily using Google ADK (Agent Development Kit).

• Context Engineering & RAG: Build and optimize end-to-end RAG pipelines. Implement advanced techniques such as smart document chunking, hybrid search, re-ranking, and dynamic context window management.

• Prompt Engineering: Develop, test, and version-control complex system prompts to ensure structured, reliable, and hallucination-free outputs from foundation models.

• Evaluation: Familiarity with LLM evaluation frameworks for systematically measuring hallucination rates, precision, and recall.

• Observability: Strong understanding of Multi-Agents Observability to debug & improve accuracy/latency

Backend Engineering

• API Development: Architect and develop high-performance, asynchronous microservices and RESTful APIs using FastAPI to connect our AI agents with the front-end and internal systems.

• System Integration: Safely and efficiently integrate external tools, enterprise APIs, and data sources into the agentic workflows so our assistants can take real-world actions.

Cloud Infrastructure & Database Management

• Vector & Relational Databases: Set up, manage, and optimize vector search capabilities using Cloud SQL for PostgreSQL (pgvector) to handle both structured relational data and high-dimensional semantic retrieval.

• Deployment & Scaling: Containerize AI applications using Docker and deploy them onto Google Kubernetes Engine (GKE).

• CI/CD & Version Control: Maintain clean code repositories using Git to streamline testing and deployment.

Required Qualifications

• Experience: 3–5 years of professional software engineering experience, with at least 1–2 years actively building and deploying production LLM or machine learning applications.

• Programming: Expert-level proficiency in Python (including a strong understanding of AsyncIO).

• Web Frameworks: Strong hands-on experience building robust backend services with FastAPI.

• AI/LLM Ecosystem: Deep understanding of RAG architectures, vector embeddings, semantic search, and prompt engineering. Active experience with Google ADK is highly preferred.

• Database Experience: Strong SQL skills and experience managing embeddings using PostgreSQL and the pgvector extension.

• Cloud & DevOps: Proven experience with GCP, containerization (Docker), and orchestration (Kubernetes/GKE).

• Software Engineering Fundamentals: Solid grasp of Git workflows, unit/integration testing (e.g., pytest), API design, and distributed system architecture.

Preferred Qualifications (Nice-to-Haves)

• Familiarity with other GenAI orchestration frameworks to bring diverse architectural perspectives and best practices.

• Experience handling streaming responses (Server-Sent Events or WebSockets) for real-time, low-latency chatbot interactions.

• Understanding of data privacy, security guardrails, and PII handling in generative AI applications.

This posting is for an existing vacancy with the organization.

AI may be used to screen, assess or select applicants for the position.