Jeremy
Perera
Associate Software Engineer at IFS. I turn ambiguous problems into clean, reliable software.
Across web, backend, and applied machine learning, with a research streak in federated learning and a habit of shipping.
- Role
- Assoc. Software Eng · IFS
- Based in
- Colombo, Sri Lanka
- Focus
- Web · Backend · ML
- Status
- Open to collaboration
A developer with a builder’s mindset.
I’m a software engineer based in Sri Lanka, currently shipping enterprise-grade software as an Associate Software Engineer at IFS. I hold a BEng in Software Engineering from the Informatics Institute of Technology, affiliated with the University of Westminster, UK.
Outside of code I’ve spent years debating competitively, including a quarter-final at the Cambridge University British Parliamentary Championship, and leading community projects with Rotaract. I bring that same energy into every team I’m part of.
Engineer at heart
I care about clean architecture, readable code and the small details that make software feel solid.
Curious by default
From web apps to neural networks in R, I love wandering across the stack and learning along the way.
Team-first mindset
Years of debating and leading clubs taught me that great work is almost always a conversation.
Where I’ve been building things.
Associate Software Engineer
- ·Contributing to enterprise software products used by customers around the world.
- ·Working across the stack: building features, debugging real-world issues and collaborating across distributed teams.
- ·Continuously sharpening engineering fundamentals: code quality, testing and design.
Intern
- ·Engaged with customers across phone, email and in-person channels to provide accurate information.
- ·Partnered with internal teams to analyse and resolve customer issues efficiently.
- ·Worked on bank-issued endorsements end-to-end.
The toolkit I bring to a problem.
A mix of languages, frameworks and habits picked up through coursework, projects and shipping real software.
Languages
Frameworks & Tools
ML & Data
Practices
Strengths
Things I’ve built and shipped.
Projects from university, collaborative work, competitions and side experiments, spanning web, mobile and machine learning.
Federated Learning for Personal Finance Prediction
Designed and implemented a complete Federated Learning system to predict disposable income from real-world finance data. Built custom FedAvg, FedProx and an original SmartFedProx strategy, featuring adaptive μ, divergence-aware client selection and server-side μ tuning. Delivered both a bespoke simulation framework and a parallel Flower (flwr) implementation, with a FastAPI dashboard, WandB experiment tracking, CI/CD and AWS EC2 deployment.
Tea Factory Ops: Sri Lanka Green-Leaf Supply Chain Platform
↗Two-person SaaS project digitising Sri Lanka's green-leaf tea supply chain, replacing entirely paper-based factory operations with a web ERP and companion mobile app. Covers daily leaf intake, supplier management, automatic monthly payment calculation, production tracking and quality assessment. Built as a full-stack monorepo; Phase 2 targets an Uber-style two-sided marketplace for suppliers and factories.
AI Running Tool: Scalable SaaS
↗A scalable SaaS platform built to help runners train smarter using AI-driven insights. Architected from the ground up for multi-tenant scalability, with a focus on clean, extensible infrastructure that can grow from a single user to thousands.
Resume Builder: Scalable SaaS
↗A scalable SaaS product for building and managing professional resumes, designed with a multi-tenant architecture and a modern, extensible stack to support growth from individual users to a full customer base.
The Predictor: AI Stock Forecasting App
AI-powered web application for stock price prediction. Led the team and contributed to the data science component. Selected to participate in Microsoft Imagine Cup 2023.
MLP Neural Networks & Partitioning Clustering
Energy-forecasting research project in R: pre-processed a real-world dataset, performed partitioning clustering and built an MLP neural network for prediction.
Appointment Application
Desktop-style doctor appointment system built in Java, with a clean Java Swing GUI and structured OOP design.
HTTP & Java Server
Built a small HTTP server and a Java client that talks to it through a local server port, a hands-on dive into networking fundamentals.
Hotel Booking Application
Hotel reservation system implemented in Java with a focus on clean state management and a simple, usable interface.
Personal Website Projects
A series of personal site experiments hosted on Firebase, built with vanilla HTML, CSS and JavaScript.
The Art of Coffee
Group coursework: a small e-commerce-style coffee ordering site with an add-to-cart flow, built in HTML, CSS and JavaScript.
The Score: Cricket Scoring App
Cricket scoring application built in Python as part of the college team, advancing to the competition quarter-finals.
Foundations in engineering, sharpened by community.
Informatics Institute of Technology
2021 – PresentSt. Peter's College, Colombo 04
2017 – 2020Beyond the classroom
Vice President (2022 – Present). Quarter Finalist at the Cambridge University British Parliamentary Debating Championship (2022).
Joint Chairperson of Sally's Manor (2022) and Treasurer of the Eco Brick Project (2022).
Senior Prefect (2019 – 2020) and Deputy House Captain (2019). President of YCS, Colombo South Deanery.
Project Co-ordinator (2019 – 2020): organised the College ICT day and supported students entering IT competitions.
Building for the community, not the résumé.
Outside of work and university, I give my time to projects that matter to the people around me. Giving back through code is one of the most rewarding things I do.
Contributing to projects I actually use.
Giving back to the tools and communities that make software better. Here’s where I’m actively contributing.
The model-definition framework for state-of-the-art machine learning, spanning text, vision, audio and multimodal models for both inference and training. Actively contributing code and reviewing pull requests on one of the most widely used ML libraries in the world.
A workflow orchestration framework for building resilient data pipelines in Python, making it straightforward to add scheduling, retries, caching and event-driven automation to any script.
A feature-rich macOS menu bar system monitor covering CPU, GPU, memory, disk, network, battery, sensors and Bluetooth devices. Contributing to one of the most popular open-source macOS utilities, written primarily in Swift.
Turns a Strava data export into an interactive HTML heatmap with six switchable layers: frequency (linear & log), pace, heart rate and gradient (absolute & directional). No API key needed, just your Strava zip export.
Got an idea, a role, or just want to say hi?
I’m always open to interesting conversations, about software, side projects, debating, or coffee. Pick whichever channel works for you.