Professional Experience
Cloudflare
Dec 2021 - Feb 2025
- Software Engineer II
- Legal Blocks Platform: Architected and implemented the Legal Blocks API, managing thousands of daily legal blocks via multi-API integrations. Developed an internal platform to streamline block creation, monitoring, and compliance.
- CSAM Matcher on Edge MVP: Delivered a scalable MVP using Workers, Vectorize, D1, and Queues, demonstrating the capability to process 500k images/second and enable global CSAM detection at the edge.
- Geo-Blocks Integration: Developed Geo-Blocks functionality within the Legal Blocks API, enabling compliance with regional legal requirements while maintaining global content accessibility.
- Legal Response Platform: Designed a secure platform for processing legal and customer data requests, simplifying complex data structures across thousands of accounts and providing secure download links to customers.
- Abuse Complaint API: Redefined the API to support action-based workflows, improving abuse report processing efficiency and integrating an auditable log system for enhanced transparency.
- Kubernetes Migration: Led the migration of nine Kubernetes microservices to a new data centre with near-zero downtime, increasing reliability and service availability.
- Block Review Requests: Enabled dashboard-driven review requests for legal blocks, enhancing transparency and satisfaction for tens of thousands of users.
- Tech Stack: Go, PostgreSQL, Kubernetes, Terraform, Cloudflare Workers, Access, Queues, D1, R2, Redis, Docker, Rust, Helm, JavaScript, React, Vectorize
Bell Canada
Sep 2021 - Dec 2021
- Software Engineer
- Built and maintained microservices to automate workflows for network engineers
- Added feature to automate firewall access rule modifications: Saved 700 hours every month
- Linked tickets from outdated ticketing software to new automation microservice, saving 100+ hours manual labor per month
Bell Canada
Apr 2020 - Aug 2021
- Software Engineer, Intern
- Implemented an hourly data reporting service to send daily statistics to 10+ management stakeholders
- Automated Cisco ASA firewall upgrades: Saved 250 hours of manual work every month
- Leveraged knowledge in Python, Flask, PostgreSQL, OpenShift, GitLab for CI/CD
Carleton University
Sept 2019 - Present
- Lead Teaching Assistant
- Lead TA for “Programming and Problem Solving with C”
- Managed lab tutorials for 150+ students each week, and provided personalized feedback
- Nominated for Outstanding TA Award 2020 and 2021 (Top 5% out of 2000+)
RISE Research Group
Sept 2019 - Present
- Research Assistant
- Experience management research using Machine Learning models
- Research jointly funded by MITACS + RISE (16k CAD funding)
Skills
- Programming: Python, Go, Rust, JavaScript, React
- Cloud: Docker, Kubernetes, Terraform, Cloudflare Workers, gRPC
- Database: PostgreSQL, Kafka, Redis, Vectorize, Cloudflare R2, Cloudflare D1
Personal Projects
Serverless Search Engine - Go
- Orchestrated 8+ independent services using Cloudflare Workers and Cloudflare Queues.
- Leveraged Algolia for semantic search, indexing, and autocomplete.
- PlanetScale for graph storage and Cloudflare R2 for webpage storage.
- Breadth-First PageRank Algorithm for distributed ranking.
- Check architecture at: https://github.com/jinayshah7/serverlessSearchEngine
The Notes Project
- Published all my 15+ detailed Machine Learning notes on GitHub
- Topics Covered: Encoder-Decoder and Attention Models, Transformers and BERT, YOLO V3, Image Segmentation using UNets, Semantic Search using Elastic Stack, Spark & MLLib, Foundations of Supervised, Unsupervised models, Deep Learning models, Foundations of Statistics, Probability, Linear Algebra, ML Interview Questions, etc.
Hand Gesture Recognition
- Trained MobileNet to classify 25 unique hand gestures solely using webcam feed
- Achieved 20x boost in live classification speed compared to InceptionV3
- 30 FPS live classification, without GPU, on dual core laptop
- Utilized: Keras, OpenCV for capturing webcam feed, PyAutoGUI for creating keystrokes and mouse events
Education
Master of IT: Data Science
Sep 2021
- Thesis: Behaviours as Vectors
- The system uses deep learning to summarize behaviour data of a player into a single vector. It’s an equivalent of word2vec for behaviours. Used OpenDOTA API to compile dataset of 700k games. Developed a framework to test vector quality and tried 270 experiments to explore the best hyper-parameters for vector quality
Bachelor of Technology: IT
May 2019
- Gujarat University (GPA: 8.0/10)
Achievements
- 6000 CAD - MITACS Research and Training Award (2020)
- Judge @ Oakhacks Hackathon, Canada (2020)
- Cost of Living Analysis Using Crowd-Sourced Data, Research Project @ Defence Canada (2020)
- Top 2 @ Canadian Partnership Against Cancer Hackathon (2019)
- 4000 CAD - IT Department Graduate Scholarship (2019)