Ena Rajović

Product Engineer.

Building climate tech at Root Global.

About me

I'm curious about complex systems. It started with atmospheric science and geography. When I discovered programming, I went from studying to building them — and learned that making them simple is the hardest part.


Back in 2019, as a second-year undergraduate, I joined a research team developing an R package — I wrote my first lines of code in R, kept going with Python and Linux, did a 6-month Le Wagon bootcamp after graduating, landed my first job as a fullstack developer, and have been building since.


At the end of 2023, when LLMs became impossible to ignore, I went independent to see for myself — and built Trash Tutor, a mobile app that uses camera input and LLMs to help with waste sorting.


At Root Global, I build end to end the products that help measure and reduce agricultural emissions — a good match for someone who studied the atmosphere and ended up writing software.

Picture of me

Experience

I worked 2 years as a full-stack developer at KURABU, a Berlin-based SaaS startup, collaborating closely with designers and developers. I led the transition from Vanilla JS to React/TS and introduced Cypress testing.

Curious about AI, I decided to leave my job and concentrate on learning new technologies. I've developed Trash Tutor for 3 months, using React Native, Cloudflare Workers and OpenAI's API.

My skills include:
  • TypeScript, JavaScript, React
  • Node.js, Cypress, Next.js
  • Elixir, Phoenix, PostgreSQL
  • Cloudflare, Heroku

Things I've built

Featured project

Trash Tutor

An AI-powered app for efficient waste sorting. Snap a photo of an item and learn how to correctly sort it based on your local recycling guidelines. Discover the appropriate disposal category, along with a clear explanation of its classification.

  • React/Typescript
  • React Native/Expo
  • Node.js
  • OpenAI API
  • Cloudflare Workers
Trash Tutor logo

Featured project

Socrates

A RAG-based technical interviewer that challenges your knowledge applying Socratic questioning - built at the Mistral AI Hackathon 2024 in Paris

  • Python
  • StreamLit
  • Mistral AI API
  • LangChain
  • Groq
Socrates logo

Featured project

REfood

A demo web app serving as a virtual marketplace for purchasing food items close to their best before date. Designed to minimize food waste, the platform offers a simple and engaging user experience for browsing and purchasing.

  • Ruby on Rails
  • Javascript
  • Active Record
  • Heroku
REfood logo

Open-source contribution

OpenFreeMap

A open source project providing free map hosting so you can display custom maps on your website and apps.

OpenFreeMap logo

Open-source contribution

RLumCarlo

A collection of functions to simulate luminescence production in dosimetric materials using Monte Carlo methods.

RLumCarlo logo

Things I've written

When I find the time, I share my learning journey on my blog.

Featured post

Less is more - even when prompting

Key takeaways from crafting the final Trash Tutor prompt that significantly enhanced my LLM responses.

  • LLMs
  • Prompt engineering
  • Function calling
  • JSON mode

Get in touch

Although I'm not currently seeking new opportunities, I love hearing about new projects and connecting with fellow developers. Feel free to reach out for an idea exchange or collaborations — I'll do my best to respond!
You can find more of my work on GitHub and LinkedIn.