Ganesh Kumar Thirunavukkarasu Kuppusami (T K)
ganeshkumartk@outlook.com
INDIANAPOLIS, IN
US
Objective
Find a job where I can do more work in public, collaborate directly with other individuals or companies, and spread inclusivity & accessibility in the world.
Education
IUPUI, Indianapolis, US
Aug 2022-Present | Masters in Human-Computer Interaction
IIITDM, Kancheepuram, India
Aug 2017-Sep 2021 | Bachelor of Technology in Smart Manufacturing
Technologies
Software
| Web | React, Gatsby, Next.js, Node.js, TypeScript, Svelte, GraphQL, Express |
| Mobile | Kotlin, Dart, Flutter, React Native |
| Machine Learning | Keras, TensorFlow, PyTorch, Hugging Face, Pandas |
| Others | HTML5, CSS3, Tailwind CSS, PWA, Bash, Groovy, JSON, Markdown, YAML, C |
Tools
| Version Control | Git, SVN |
| Continuous Integration | Travis CI, GitHub Actions |
| Cloud | AWS, Azure, Google Cloud Functions |
| Hosting | Netlify, Vercel, Heroku, Github Pages, Firebase |
| Databases | MongoDB, MySQL, Firestore, FaunaDB |
| Development Lifecycle | JIRA |
| Design | Figma, Miro, Adobe XD |
Employment History
India Health Link | Sr. Software Engineer
June 2021 - October 2021
- Worked as lead in Health Journal feature implementation
- Conducted brief feasibility study for Food journal and Activity journal in mobile application.
- Design and developed Screens for Food journal in IHL Care app along with Mobile app team.
India Health Link | SDE Intern
June 2020 - October 2020
- Joined as an Summer Intern in Mobile Application Development Team at IHL.
- Worked on Cross Platform mobile App development in Flutter.
- Handled UI design, API Integration, Database management.
Talks & Appearances
08/04/2019 | Program Alexa with Python (pyalexa)
06/25/2021 | Tensorflow in Microcontrollers
Projects
Academia
- single image to 3D object mesh.
- detects 36 objects trained(currently) from ShapeNET (Stanford).
- leverages GraphCNN to spatial degenerate objects from image.
- tracks possible blindspot using G-ResNet.
- built with Flutter, PyTorch, FastAPI and 3d printable.
- classifies images of pills and identifies for quick medication.
- detects 12 set of pills using PILLNET.
- works entirely on your device (on-device ML).
- built with React and tensorflow.js.
Personal
- social app to request and deliver kindness in concept of letters and mailboxes.
- totally anonymous.
- spam & bad word filters, reporting and in-built moderation.
- delete everything, anytime & no ads or data selling.
- Built with Flutter & Firebase.
- web app to find Supermarket waiting times during Pandemic.
- using Traffic and popularity times API of HERE maps
- works based algorithmic detection on footfall times.
- built with React, HERE Map API and Leaflet.
Packages / Extensions / Bots
💡 hue-sdk
- Philips Hue JavaScript SDK.
- Written in TypeScript.
- Powers functionality in my Twitch bot.
- Over 1,500 downloads!
🧑🏼💼 baasha
- Command line tool for object segmentation.
- Built with JavaScript and Tensorflow.js.
- 500 downloads/week!