this is really great advice. did crossfit for a bit and stopped, but some of the mobility stretches using a lacrosse ball and band have changed my life. (warm up first)
Avant (fmrly DebtEye YC S2011) | Los Angeles, Chicago | Backend, Frontend, Mobile, Analytics/Machine Learning | Full-Time
Ruby, R, Python, Redshift, Postgres, Scala
Avant Credit Corp ('Avant') is a fast-growing startup that offers tremendous advancement opportunities to its employees. Since its launch in December 2012, Avant, a consumer finance company, has changed the way customers borrow money -- providing them with a new and unique online experience. Avant uses state-of-the-art analytics methods in order to offer its lowest possible interest rates to the widest range of customers. Since inception, Avant has secured over $300 million in equity financing and over $700 million in debt financing to continue to grow at an exponential pace. Previously, Avant's founders created the leading international online short-term lender currently employing over 1,000 individuals and generating over $700 million in annual revenue. Avant’s founders have also launched an innovative real estate investment trust that provides affordable housing options in Chicago, Indianapolis and Baltimore with over 10,000 apartments owned and managed.
ReadyForZero (YC S2010) | Los Angeles, CA | Backend, Frontend, Mobile, Analytics/Machine Learning | Full-Time
Clojure, ClojureScript, R
ReadyForZero's award winning product and blog helps consumers get out of debt. We've helped hundreds of thousands of people pay off >$200M in debt. We were recently acquired by Avant, a Chicago-based online lender to offer our product to their borrowers and are hiring aggressively in Los Angeles.
Author here, thanks for the comment -- since that seemed to have been lost, I'm using "fundamental" because transducers let you describe your logic so that it that can be applied over sequences and non-sequences in a way that you cannot by just applying composed logic functions through existing clojure machinery.
Probably way beyond the scope of that project, but actually this is a pretty interesting and well-studied problem called segmentation, where you train a system to insert the correct breaks. This is used in natural language problems like OCR or machine translation.
Maybe google "machine translation segmentation" or OCR for references to papers on the topic, but methods for doing this are really successful (especially on english).
It took me awhile to get, but my aha moment was understanding its relationship to regular PaaS's (eg: heroku).
What more could you do if you didn't need to deploy your entire stack with each little change you make? If you could only deploy your frontend (in different versions simultaneously) pointing to a backend that's always up, it enables much faster iteration: testing, bugfixing, demoing things that otherwise you might not deploy different versions for because it's too painful.
For example, you could A/B test meaningful portions of frontend code in a more scalable way than, say, Optimizely. Or you might point a user that's having problems to a specific build (which you can deploy instantly) and see if it fixes their problem _on their live account_. Or you can hack together a new feature that works _on a live account_ for a presentation, and push it without worrying about mucking up or interrupting production traffic. Or turn off minification instantly and deploy that.
Plus, you don't have to worry about refining internal frontend build scripts, it's plug and play.