I want to share some of the tools that power my work as a data science student in and out of university.
Lets start with hardware. I've used windows machines for as long as i've had a laptop throughout university. I decided it was time to try something new and dive into the macOS ecosystem. I’ve always cared about having a quality laptop, so when the MacBook Pro M4 dropped(like yesterday), it felt like the perfect chance to upgrade.
I liked the design, great battery life, and performance bump, all of which was what I was looking for. So far i've enjoyed its performance emmensely. This included mini LLM training, matrix operations data processing, and utilizing Apple's MLX framework. Usually heavy machine learning work I do is using company dedicated cloud compute resources on AWS or GCP.
Update: I later purchased a Mac Mini M4 to compliment my laptop. It serves as a stationary setup for my desk. It’s been perfect for keeping my workspace organized—I can leave it running at home for heavier data science tasks while taking my MacBook to office or coffee shops.The seamless integration between the two is a dream; I can start a project on one device and pick it up on the other without missing a beat.
Lets talk about software. I’ve always believed productivity starts with a clean workspace—both physical and digital. For coding, I split my time between Visual Studio Code and Zed. VS Code has a wide variety of extensions even for data science while Zed is more for speed and a enjoyable minimalist design. Sometimes its nice to use a minimalist editor such as Helix to focus on rapid edits.
For organization I use Notion. I dump everything here such as univeristy lecture notes, personal project ideas, work/meeting notes. I enjoy its user friendliness and natural typography and annotations.
I also use Obsidian, where I connect the dots between ideas. Its markdown-based vaults and graph view can help with mind maps and its simple minalist design keeps me focused with no distractions.
Some bonuses. I use iTerm2 as my terminal of choice since it offers many community made extensions and has good customization and personalization. For API testing I like HTTPie. It replaces Postman, offering a more light weight and simple interface. It also works offline and offers a nice CLI interface!
I honestly dont use THAT many tools and all my choices are my preferences. But feel free to take some of my recommendations :)