Building AI‑powered
systems at
Canada's largest bank.

Royal Bank of Canada · Toronto, ON

• Architect end-to-end AI call center platform integrating SIP telephony, real-time speech-to-text transcription, and LLM summarization pipelines processing thousands of daily banking interactions.
• Engineered LLM ensemble labelling pipeline using GPT-4o mini, GPT-4.1 mini, and GPT-5 mini with consensus voting and tie-breaking escalation to larger thinking models; generated high-quality training labels at scale for fine-tuning.
• Implemented knowledge distillation workflow: used LLM ensemble as teacher to fine-tune Arctic Embed 2.0 Large as domain-specific student model for banking intent semantics.
• Built continuous training MLOps pipeline on S3 + Apache Airflow; automated dataset ingestion, model retraining, evaluation, and deployment to OpenShift.
• Benchmarked SVM (RBF kernel), logistic regression, SetFit (head and full fine-tune), centroid, and fine-tuned Arctic Embed classifiers for production intent routing.
• Led BART intent classification system using Jina v3 / multilingual-e5 with ONNX Runtime; reduced inference latency from 8 s → 50 ms on CPU-only OpenShift 4.
• Built Spring Boot microservices with k-NN routing of 14,000+ banking intents across 110 workflow categories; achieved 95% semantic naming accuracy via LLM-assisted clustering.

Royal Bank of Canada · Toronto, ON

• Built centralized vulnerability triage platform spanning Canada, US, and UK subsidiaries; reduced security assessment time by 60% across multi-subsidiary operations.
• Designed data ingestion pipelines integrating Snyk, NexusIQ, Kenna, Tenable, and Aqua security feeds into unified Snowflake data warehouse.
• Delivered full-stack platform using React, FastAPI, Snowflake, Redis, and SQL Server deployed on OpenShift.

Royal Bank of Canada · Toronto, ON

• Automated large-scale chaos experiments on VMs and Kubernetes namespaces across AKS, OpenShift, and VMware using Tanium and Gremlin APIs; reduced experiment setup time by 90%.
• Architected SRE data enrichment pipelines integrating PagerDuty, ServiceNow, and ELK Stack; reduced mean time to detect (MTTD) by 30%.
• Built observability data pipelines using Python, Pandas, Oracle DB, SQL Server, and PostgreSQL to feed critical internal monitoring dashboards.

Royal Bank of Canada · Toronto, ON

• Architected mTLS API gateway connecting Slack applications to internal banking services using Golang, HashiCorp Vault, GitHub API, and Nginx on OpenShift; enabled enterprise ChatOps saving hundreds of engineer-hours weekly.
• Built bulk employee onboarding service reducing provisioning time from hours to minutes using Golang, Gin, PingFederate SSO, and ServiceNow on OpenShift.
• Developed metrics pipeline tracking open-source/innersource artifact reuse across the organization; delivered executive dashboards using Logstash, Kibana, Redis, NexusIQ, and Artifactory.
• Created innersource SDKs in Python/Golang reducing Slackbot onboarding from 2 weeks to 2 days for 15+ teams.

Royal Bank of Canada · Toronto, ON

• Built internal Stack Overflow-style Q&A platform with gamification; increased answered-question rate by 25% and connected developers across 10+ lines of business.
• Architected production-ready WebSocket Slackbot pub-sub framework scaling horizontally without message duplication using Golang and RabbitMQ.

Royal Bank of Canada · Toronto, ON

• Implemented ChatOps into DevOps pipeline via scalable Slackbot saving 20+ engineer-hours per microservice per week using Python, MongoDB, Redis, RASA NLU, Docker, OpenShift, and Kubernetes.
• Built ML-based SRE pipeline improving application resilience and reducing toil/recovery times using Elasticsearch, Logstash, Kafka, and IBM CDP.
• Developed automation test portal reducing test execution time by 80%, saving $100k+ annually.
• Built intelligent war room using Raspberry Pi, Python, Snips, PagerDuty, Moogsoft, and ServiceNow to surface solution vectors from similar past incidents.

University of Toronto · Remote

• Developed web application for the History Department to track course enrollment trends; built with Django, Bootstrap, Redis, jQuery, Nginx, Chart.js, and PostgreSQL.