AI systems
for RBC's
call centre.

Royal Bank of Canada · Toronto, ON

• Own the intent detection and classification models for the bank's call-centre platform: labelling, training, and inference over live call transcripts from AWS Transcribe, across thousands of daily banking calls.
• Built an LLM ensemble labelling pipeline (GPT-4o mini, GPT-4.1 mini, GPT-5 mini) with consensus voting and escalation to larger thinking models for tie-breaks, to generate training labels for fine-tuning.
• Ran a knowledge distillation workflow using the ensemble as teacher to fine-tune Arctic Embed 2.0 Large into a domain student model for banking intent semantics.
• Built the continuous training pipeline on S3 and Apache Airflow: dataset ingestion, retraining, evaluation, and deployment to OpenShift.
• Benchmarked SVM (RBF), logistic regression, SetFit (head and full fine-tune), centroid, and fine-tuned Arctic Embed classifiers for production intent classification.
• Served Arctic Embed 2.0 Large with ONNX Runtime on CPU-only OpenShift, tuned with AVX-512 and related flags (no VNNI), inside a Spring Boot service on Java 25 that handles thousands of concurrent advisor and client conversations in real time.
• Trained a class-balanced logistic-regression head on the ensemble-labelled data to classify caller intent live as each conversation progresses, reaching up to 85% accuracy.

Royal Bank of Canada · Toronto, ON

• Built a centralized web platform for vulnerability triage and exemption services across RBC subsidiaries in Canada, the US, and the UK.
• Built data ingestion pipelines pulling Snyk, NexusIQ, Kenna, Tenable, and Aqua feeds into a single Snowflake warehouse.
• Delivered the full stack: React and Tailwind front end, FastAPI services, 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 the Tanium and Gremlin APIs, cutting experiment setup time by 90%.
• Built SRE data enrichment pipelines across PagerDuty, ServiceNow, and the ELK Stack, cutting mean time to detect by 30%.
• Built observability data pipelines in Python and Pandas over Oracle, SQL Server, and PostgreSQL to feed internal monitoring dashboards.

Royal Bank of Canada · Toronto, ON

• Built an mTLS API gateway connecting Slack apps to internal banking services using Go, Apache APISIX, HashiCorp Vault, Apigee, the GitHub API, Elasticsearch, and Nginx on OpenShift. It enabled ChatOps workflows that saved teams hundreds of hours a week.
• Built a bulk employee onboarding service that cut provisioning from hours to minutes using Go, Gin, PingFederate SSO, and ServiceNow on OpenShift.
• Built a metrics pipeline tracking open-source and innersource artifact reuse across the org, with executive dashboards on Logstash, Kibana, Elasticsearch, Redis, and NexusIQ.
• Wrote innersource SDKs and templates in Python and Go that cut Slackbot onboarding from two weeks to two days for 15+ teams.

Royal Bank of Canada · Toronto, ON

• Built an internal Stack Overflow style Q&A platform with gamification. It lifted the answered-question rate by 25% and connected developers across different lines of business.
• Built a production WebSocket Slackbot pub-sub framework in Go and RabbitMQ that scales horizontally without duplicating messages.

Royal Bank of Canada · Toronto, ON

• Built a Slackbot ChatOps pipeline that saved 20+ engineer-hours per microservice each week using Python, MongoDB, Redis, Rasa NLU, Docker, and OpenShift.
• Built an ML-based SRE pipeline that improved app resilience and cut recovery toil using Elasticsearch, Logstash, Kafka, and IBM CDP.
• Built an automation test portal that cut test execution time by 80%, saving over $100k a year.
• Built a private incident war room on a Raspberry Pi with Python, Snips, PagerDuty, Moogsoft, and ServiceNow that surfaced fixes from similar past incidents.

University of Toronto · Remote

• Built a web app for the History Department to track course enrollment trends using Django, Bootstrap, Redis, jQuery, Nginx, Chart.js, and PostgreSQL.