Related Posts
Need some advice here. I am a fullstack developer with 5 yoe in Angular and Python. My aim is to crack FAANG companies.Now I got an offer from HSBC in a credit risk model monitoring role using Python.It is close to a data engineer role.
My question is that will it be a good idea to shift from development role to a model monitoring role if I want to move to FAANG in the future?Or does FAANG not prefer people who are not in core development roles?Amazon Microsoft Google Adobe PwC EY Citi Barclays JPMorgan Chase
More Posts
Additional Posts in Data & Analytics Consultants
Got messaged by a C3 . ai recruiter. Read that wlb is bad and that the interview process is absurdly long, but the Glassdoor reviews are 4.2 and can't find actual hours worked posted by anyone. How's the culture really? I'd be aiming for DS consulting, something more functional but with DS/ML concepts as my differentiator.
C3.ai, Inc.
New to Fishbowl?
unlock all discussions on Fishbowl.




Bowl Leader
That's one of those generic descriptions. Don't sweat it. Review your data mining class if you had one ("classic" ML) and you'll mostly be ok. Since you're at a bank, brush up on time series stuff just in case.
A typical ops/applied science interview will have you go over ML at a basic level and then ask you to discuss your specialty. You may be asked to derive OLS, build naïve Bayes from scratch, and draw out gradient descent math. Beyond those, it'll be conceptual stuff (explain boosting, what's a convolution, what are transformers, etc). If you're asked to implement an LSTM on the spot, your interviewer is playing hard ball--unless you're a PhD with LSTM in the name of the dissertation on your resume, of course...
Kaggled does seem like a good way to get some modeling practice, but don't forget the domain knowledge and all the production stuff. I'm talking about understanding where a model can fail, catching feedback loops, recognizing how to match a model to a business requirement, etc. Finding survey papers and blogs on things like tech debt in ML shouldn't be too hard, and will really set you apart from the avalanche of data science new grads who are also all Kaggling.
Thank you! Appreciate it
It’s really true what they say about projects providing the best way to learn those skills. I start on Kaggle to get basic ideas for a project, then build off of them to make them more customized and relevant to the concepts/areas you want to focus on
Thank you!