# Kagi ML dev challenge Welcome to your Kagi demo project! A few guidelines: - The tasks can be solved in various ways; do it in a way that will show off your skills. - Feel free to ask for clarifications - The deadline for submission is two weeks or sooner - The tasks can be solved in a variety of ways, do it in a way that will show off your skills. - Feel free to make reasonable assumptions or ask clarifying questions - Deliverable is a Github repo and a direct web link to try the solution. - You have one chance to leave first impression - try to build something that is intuitive, easy to run and has readable code Thank you for participating. Good luck, and have fun! ## Warmup task - Reranking Hacker News Build an API that given user submitted bio, returns top 500 stories from [Hacker News](https://news.ycombinator.com) front page (use their API for this), ranked in the order of relevancy to user's interests. Example input: I am a theoretical biologist, interested in disease ecology. My tools are R, clojure , compartmentalism disease modeling, and statistical GAM models, using a variety of data layers (geophysical, reconstructions, climate, biodiversity, land use). Besides that I am interested in tech applied to the a subset of the current problems of the world (agriculture / biodiversity / conservation / forecasting), development of third world countries and AI, large language models. Output: Top 500 items from Hacker News, ranked in the order of relevance to the user. The solution should optimize for development time - what is the simplest/quickest way to do this that you can think of? Spend no more than 4 hours on this. ### Deliverables - GitHub Repo - Link to a live demo containing a text area for the input and a button to submit. The results will be shown below. - Explain your thought process and decisions you made during creation of the solution. ## Main task - Query reformulation ML model Build an API to a machine learning model that given an input query can generate one or more adequate search engine queries to obtain requested information. Your API should have **maximum 100ms latency** on a consumer grade CPU. Examples of good API outputs for given inputs: - In what year was the winner of the 44th edition of the Miss World competition born? 44th Miss World competition winner birth year - Who lived longer, Nikola Tesla or Milutin Milankovic? Nikola Tesla lifespan Milutin Milankovic lifespan - Author David Chanoff has collaborated with a U.S. Navy admiral who served as the ambassador to the United Kingdom under which President? David Chanoff U.S. Navy admiral collaboration U.S. Navy admiral ambassador to United Kingdom U.S. President during U.S. Navy admiral's ambassadorship - Create a table for top noise cancelling headphones that are not expensive top noise cancelling headphones under $100 top noise cancelling headphones $100 - $200 best budget noise cancelling headphones noise cancelling headphones reviews - what are some ways to do fast query reformulation fast query reformulation techniques query reformulation algorithms query expansion methods query rewriting approaches query refinement strategies ### Deliverables - GitHub Repo - Link to a live demo containing an input field. The results will be shown below. - Explain your thought process and decisions you made during creation of the solution.