Machine Learning Engineer at Quickchat

Posted on: 04/21/2022

Location: (REMOTE)

full time

Original Source

Tags: django nginx blockchain javascript react ml python

Join our Technology team to build consumer products and business solutions on top of state of the art language models like OpenAI’s GPT-3. From day one, you will take ownership and contribute to key parts of our technology. We live and breathe Machine Learning. We love Python, Linux and Git. We tolerate JavaScript, CSS, even React Native at times. ### Requirements: * 1+ years in Software Engineer or ML Engineer roles * deep understanding of Machine Learning theory and practice * technical fluency in English * great communication skills ### Bonus points for: * Kaggle competitions (walk me through your solution!) * interesting side projects (we will want to ask a lot of questions about them!) * deeply technical interests other than software / Machine Learning * currently live coding projects (we will want to have a look at them before your interview) * teaching us something we didn’t know before about our product / industry * ever having written AWK scripts * experience with Django, single-page applications, Nginx, cloud computing, API design * experience deploying apps to the App Store or Google Play * deep understanding of and experience working with GPT-3 and other language models * experience training ML models for time-series data * experience training classification models for extremely imbalanced classes * reading machine learning papers in your spare time (share your favourites!) * telling us about something that’s wrong with our product along with a concrete and detailed plan on how you would try to improve it * deep first-principles understanding of some computing process, e.g. what exactly happens when you type a URL and press enter, exactly how blockchain / a neural network / a chess-playing algorithm / Google’s Ad bidding / Shazam’s song search works (we will definitely want you to tell us all about it!) * a story about your ML model that worked great on historical data and failed in production * a story about how you spent hours labelling data by hand for your ML model * a story about how you spent 3 days debugging something and finally found a solution * a story about how you had no idea how your users were using your product until you started measuring it * expertise in middle-out compression