AI Automation Engineer / Developer
Toronto, OntarioFull-TimeMid-levelAI / Data Science
What you'll be doing
- Design and implement AI automation pipelines using Python (FastAPI, LangChain, spaCy) or Node.js (LLM API wrappers, task automation frameworks).
- Build and optimize OCR-based data extraction systems for invoices, receipts, and other unstructured documents, integrating semantic search and rule-based classification workflows.
- Develop custom AI agents for reconciliation, tagging, and anomaly detection across financial and operational datasets.
- Fine-tune and orchestrate large language models (LLMs) for document understanding, summarization, and context-aware entity or workflow matching.
- Deploy scalable inference pipelines on GCP Vertex AI, AWS Lambda, or Dockerized microservices, ensuring reliability and cost efficiency.
- Integrate AI results into user-facing workflows via interactive dashboards, APIs, or automation tools, enabling seamless user adoption.
- Monitor and maintain model performance through automated evaluation frameworks, triggering retraining pipelines based on data drift or performance degradation.
- Collaborate with cross-functional teams (data, DevOps, and product) to ensure AI automation solutions align with business goals and compliance standards.
What you bring
- Bachelor’s or Master’s degree in Computer Science, Data Science, or related field.
- Proficiency in Python or Node.js, with hands-on experience in FastAPI, LangChain, or spaCy.
- Strong understanding of OCR, NLP, semantic search, and LLM-based automation.
- Experience deploying solutions on AWS, GCP, or containerized (Docker/Kubernetes) environments.
- Familiarity with vector databases (e.g., Pinecone, FAISS, Weaviate) and prompt orchestration frameworks.
- Strong problem-solving skills and ability to translate business needs into scalable technical solutions.
Nice to Have
- Experience with MLOps, Vertex AI Pipelines, or AWS SageMaker.
- Exposure to financial or document-intensive automation systems.
- Understanding of data drift detection, retraining strategies, and LLM evaluation metrics.
