Episode Summary: A few years ago I promised you a blog series on how to start your own consulting business in machine learning: getting set up, figuring out legal & intellectual property rights, finding consistent work, scoping in the face of research uncertainty, the project life cycle, mistakes to avoid...
I gave a few talks on this topic but never got around to writing that blog series. Today I sat down with John D. Cook, a fellow top ML/stats consultant for companies like Amazon, Google, Microsoft or Amgen, and we finally got to discuss (some of) that: John's background, his consulting for pharma and legal, project pricing, the best tool for the job vs. the cost of moving across tools, the Python language and its community, and more.
Of interest to people with a vicious streak of independence.
Links & resources:
John's consulting business: johndcook.com aka Singular Value Consulting.
John's instantly relatable, quirky and fun writings on his personal blog. RECOMMENDED! To get a taste, check out Don't invert that matrix, R language for programmers, Statistical distributions chart / cheat-sheet, Bayesian consulting...
Twitter: John's personal @johndcook + his 18 (!!) other thematic twitter accounts.
So you want to be a data science consultant (or hire one)? (my presentation from Berlin Buzzwords 2015) and From Research to Industry, in Ten Not-So-Easy Steps (similar presentation from the Industry panel at SIGIR 2016).
The podcast lives on SoundCloud, plus this time I submitted it to iTunes, Stitcher and YouTube as well. I hope you like it!
Poll: In our chat, we barely scratched the surface. Which side of consulting is the most interesting to you? Let me know in the poll below and I'll cover it more. Thanks![yop_poll id="9"]
Pingback: New podcast interview
Pingback: John D. Cook on Data Science Consulting [PODCAST] | StratCom
Great podcast, full of practical insights. Some of them I learnt the hard way:
– the importance of specialisation
– avoid keeping up too much on the literature and tooling
– scoping the projects with flexibility when possible, otherwise just bill on time
– as freelancers, we have to deal with marketing
Having confirmations on these topic was of great value.
I’m going to share, thank you!
And if you guys want to collaborate, I have a background on neural nets in research and had success with interactive data viz in the browser as a freelancer. I wish to fuse these two interests somehow.
Thanks Piero! Glad you found it useful 🙂
Looking back, I felt overwhelmed by the amount of information I wanted to cover and as a result, we didn’t get deep enough.
Next time, probably better to focus on a single smaller theme & go more in depth.