Show HN: I made a medical test simulator, to explore statistical errors https://ift.tt/2VMLU8e

Show HN: I made a medical test simulator, to explore statistical errors https://ift.tt/2VRF6WT April 22, 2020 at 02:54PM

Comments

Post a Comment

Popular posts from this blog

Launch HN: Epihub (YC S20) – Shopify for teaching online https://ift.tt/3l3c6XN

Launch HN: Epihub (YC S20) – Shopify for teaching online Hey HN! I’m Uday, and I co-founded Epihub [0] with Kwasi and Michael ( https://epihub.com ). Epihub is Shopify for teaching online. Our software lets you schedule, meet, and bill clients from your own website. A few years ago, we started building a product called Epigrammar, which was a collaborative document annotation tool that let teachers rapidly give feedback to their students by identifying trends in their feedback. Kwasi and I really wanted to see if we could scale the tutoring experience to an entire classroom, since my co-founder Mike was teaching Classics at both a private school in Connecticut while running a non-profit tutoring program in Latin/Greek for public school students in New York. Mike would try out our products that we had built over the weekend during the week (sometimes to success), but oftentimes, things were not actually helping him teach. That’s when we'd go back to the drawing board. We spent a few...
Read more

Launch HN: Narrator (YC S19) – a data modeling platform built on a single table https://ift.tt/34bcMD0

Launch HN: Narrator (YC S19) – a data modeling platform built on a single table Hi HN, We’re Ahmed, Cedric, Matt, and Mike from Narrator ( https://www.narrator.ai ). We’ve built a data platform that transforms all data in a data warehouse into a single 11-column data model and provides tools for analysts to quickly build any table for BI, reporting, and analysis on top of that model. Narrator initially grew out of our experience building a data platform for a team of 40 analysts and data scientists. The data warehouse, modeled as a star schema, grew to over 700 data models from 3000+ raw production tables. Every time we wanted to make a change or build a new analysis, it took forever as we had to deal with managing the complexity of these 700 different models. With all these layers of dependencies and stakeholders constantly demanding more data, we ended up making lots of mistakes (i.e. dashboard metrics not matching). These mistakes led to loss of trust and soon our stakeholders were ...
Read more