Hey, I'm Lucas
Fullstack Developer

Hi, I'm Lucas, a fullstack developer currently studying Computer Science at the University of Waterloo. I'm currently working at SlimeScholars as a Founding Engineer.

Work

Here are my most relevant work experiences. I will try to keep this updated.

SlimeScholars - Founding Engineer (May 2024 - Aug 2024)

Led backend development of a gamified education platform built with Next.js, React, MongoDB, and AWS, accumulating over 10,000 monthly users.
Created parser for custom markdown for educational content, using google analytics + in-house system to track student engagement on lessons and quizzes.
Managed a team of 5 engineers, ensuring robust and efficient code with Jest and benchmarking.
Slashed storage needs by 30% by optimizing Mongoose models, MongoDB aggregate pipelines and S3 as an object store.
Developed real-time matchmaking algorithm to gamify learning through competitions, using cron and aggregate queries to match teams of students of similar sizes and skill.

Retrospect (Tenacity Labs) - Backend Developer (Mar 2024 - April 2024)

Built an app to track memorable experiences with RESTful Go stdlib backend, using Gorilla Mux for routing, MySQL and Google Cloud Buckets as an object store and CDN.
Configured crontab and SendGrid on Google Cloud infrastructure to automate email notifications.
Designed a social recommendation algorithm using a social graph to build an effective referral program.

unfollowed.lol (Tenacity Labs) - Software Engineer (Feb 2024 - Mar 2024)

Built JavaScript browser extension that leverages Instagram’s internal client-side APIs to comply with server-side rate limits.
Streamlined Postgres queries using Django’s ORM, also cutting data redundancy by 80% through diffing techniques.
Launched web extension on Product Hunt ranking #8 daily with more than 1500 daily active users.

Projects

Here are some of the memorable or flashy projects I've worked on. I will try to update this as I work on new things.

betUFC (Dec 2023 - Jan 2024)

Regularized data and engineered features for regression algorithm with LightGBM and Flask to predict UFC fight outcomes.
Fixed class imbalance that arose from a limited dataset, bringing results that differed by around 10% to under 2%, generating 20.3% ROI over 4 months, by modifying training data
Optimized the model’s AWS deployment, reducing timeouts by 30% by multithreading and adjusting worker counts.

Marble Investments (Nov 2023 - Dec 2023)

Built full-stack web app with Next.js, React, and MongoDB for an investment club, handling transactions and newsletters.
Processed membership fees using Stripe API and managed 2FA using React Email and Resend.

Visual Sorting Algorithms (May 2023 - Aug 2023)

Animated comparisons and swaps of sorting algorithms through DOM manipulation, React hooks, and state variables.
Manipulated promises to work around React’s data binding with state variables to allow for timed animations.