Mississauga, Ontario, Canada
I am a full-stack developer with over 7 years of experience in developing applications and cloud infrastructure using AWS services. My expertise includes Typscript, Python, and the React ecosystems. Currently, I am working at BMO Financial Group as a Lead AWS Developer using AWS CDK (Cloud Development Kit) to deploy infrastructure as code solutions. Prior to that, I worked at Romet Limited as a Full Stack Developer where I developed Android and iOS native apps for gas meter field technicians, maintained cloud infrastructure, and worked on AWS IoT Core for data ingestion. Please checkout my latest open-source work: https://github.com/DominicFung Personal Website: https://www.dom-fung.com/
Implemented a CLI, integrated with git hooks, to trigger deployment of AWS CDK projects on commit using internal built pipelines on Bitbucket & GitHub. The CLI cut the number of manual steps by 67% and is integrated with multiple development environments (AWS Cloud9 & Local). Created a new EFT Platform in NextJS & React, load balanced between multiple file-transfer nodes.
Worked on Android and iOS native apps for gas meter field technicians via Java/Swift with REST calls to AWS using API Gateway and lambda micro-services. Maintained cloud infrastructure using CloudFormation and the serverless framework. Implemented multi-tiered deployment (test and production environments) with CI/CD features via CodeCommit, CodeBuild, CodePipeline and CodeDeploy. Augmented and implemented new features on the current data-portal built with Typescript React using GraphQL via AWS Amplify/Appsync talking to DynamoDB and RDS (Postgres). Worked on IoT Core and IoT Rules for data ingestion and device commission/decommission. Used React-Native to build prototype Android/iOS/Desktop apps to communicate directly with meters using serial communication (RS232).
Oslyn.io is an app looking to change the way we make music! Under the hood, she's an AI (Tensorflow) that provides real time accompaniment to musicians playing their favourite instrument. We've utilized some of the most cutting edge technology in as part of the stack! Starting with the platform of chose - We prioritized serverless architecture, lowering cost and enabling unlimited* scalability (S3, Appsync, AWS-amplify, dynamoDB). We chose to use the React ecosystem for both our app and website (React-Native/Reactjs) for code consistency and reusability. Since our app responds in real-time, we decided to write the audio layer in Java/Swift and take advantage of the native performance gains on Android and iOS platforms respectively. The AI model also resides within the native code. Speaking of AI, we utilize AWS Sagemaker as the our training platform, kicking off ML training instances via Python on Jupyter notebooks within the environment! Our original architecture utilizes Pytorch & Pytorch mobile, but due to dependency conflicts with React-native, we migrated the models to Tensorflow & TFlite. Due to the serial nature of music, we're using research on recurrent neural networks as the bases of our AI model.
Helped the Workfusion platform move beyond its proof-of-concept stage. Rapidly designed and developed a 3 tier development progression system in compliance with the BMO operating model. Standardized and implemented a 3 level production support process. Onboarded and supported multiple projects on the automation platform. Helped build a Knowledge base and tools to augment developers on the platform. Worked with the vendor, in performance and compliance related initiatives.
Nodis.io is an innovative marketing platform for local business, generating engagement through small activities and rewarding users with cryptocurrency they can later spend on stores. Checkout out our NEO smart-contract repo: https://github.com/nodis-io/nodis-token-smart-contract