Every ML Eng book and resource I've ever read recommends that any ML product should start incredibly small, starting with something that can be developed in a day or so. Then iterate on it, only pulling in new features and only updating the model whenever those prove to add predictive value.

Every ML project I've been on has started with "we know we need these 372 features and need 12-24 months to get an MVP".

#featureengineering #ml #mleng

⁉️ Trying to develop your eng skills or pivot to #mleng or #dataeng and don't have prior CS experience?

➡️ The Missing README by Chris Riccomini, Dmitriy Ryaboy ( https://read.amazon.com/kp/embed?asin=B08XM2CDZM&preview=newtab&linkCode=kpe&ref_=cm_sw_r_kb_dp_4AEZSC0KTN6PYCEVA7KF&utm_content=bufferbec7b&utm_medium=social&utm_source=linkedin.com&utm_campaign=buffer )

➡️ The Coding Career Handbook by Swyx (https://learninpublic.org/?utm_content=buffer9108b&utm_medium=social&utm_source=linkedin.com&utm_campaign=buffer )

➡️ How Computers Really Work by Matthew Justice ( https://www.amazon.com/Amps-Apps-How-Computers-Work/dp/1718500661/ref=sr_1_3?crid=3O6LN6GLFSH23&dchild=1&keywords=how+computer's+really+work&qid=1635179511&sprefix=how+computer's+really+wor,aps,150&sr=8-3&utm_content=buffer60283&utm_medium=social&utm_source=linkedin.com&utm_campaign=buffer )

➡️ MIT's Missing CS Semester: https://missing.csail.mit.edu/?utm_content=bufferdee25&utm_medium=social&utm_source=linkedin.com&utm_campaign=buffer

➡️ Beyond the Basic Stuff with Python by Al Sweigart ( https://www.amazon.com/Python-Beyond-Basics-Al-Sweigart/dp/1593279663/ref=sr_1_1?dchild=1&keywords=beyond+the+basic+stuff+with+python&qid=1635179598&sprefix=beyond+the+basic+stuff+with+pyth,aps,137&sr=8-1&utm_content=buffercfe07&utm_medium=social&utm_source=linkedin.com&utm_campaign=buffer )

👉🏻 It is a truth universally acknowledged, that a data scientist in possession of a trained model, must be in want of a reliable means of productionization and deployment.

👣 And the journey of a thousand pipelines starts with...
knowing how to appropriately package your models from the get-go. 📦

This blog post is for you: https://medium.com/kitchen-sink-data-science/software-fundamentals-for-machine-learning-series-understanding-the-why-of-vms-containers-89621cf66d23?source=friends_link&sk=4b4a9f37c1f609db2addbeb6fa219fb8&utm_content=buffer8c348&utm_medium=social&utm_source=linkedin.com&utm_campaign=buffer

#mlops #mleng #productionml #datascience #productdatascience