Artificial Intelligence is the bicycle for our Technology — My Udacity AMA
I did an AMA for the Udacity AI-Deep Learning-Data Science Community. This is the selection of the top questions and answers from the session.
Firstly, Karen Baker and Martin McGovern from Udacity help organize and facilitate this AMA for the life long learners at Udacity. I am deeply thankful to Karen, Martin and Udacity for this opportunity to share the knowledge.
Computers are like a bicycle for our minds - Steve Jobs
Artificial Intelligence is the bicycle for our technology.
Best AMA questions and answers:
QQ: What is the best piece of advice you’ve ever received in your career?
VK: I have got some good advice from books as well as mentors. One that particularly stands out from the Context of Enterprise AI (Since I work in the Enterprise):
“Always remember in the Enterprise, the AI’s goal is to solve business problems.” — Source
There is often a difference between the quantity you want to measure and the one you can measure.
When these differ your model will become good at predicting the quantity you measured, not the quantity for which it was meant to be a proxy -Yonatan Zunger
QQ: What suggestions do you have around building your portfolio?
VK: Firstly, I would suggest to carefully understand where your strengths lie. There are different verticals like NLP, computer vision, speech and so on. Don’t rush into these subdomains of the data science just because they are in high demand. Focus on your personal strengths and interests. Based on the strengths, places like Kaggle help you explore data sets which might be helpful. A good website/blog to showcase data visualizations is very important. Acing AI’s Acing Portfolios is designed to showcase that.
QQ: What are the stumbling blocks you see most AI enthusiasts fall into/What are the common hurdles and how do you recommend overcoming them? (process point of view from learning to getting their first job)
VK: That depends on what they are going after. AI in Consumer Internet Companies is very different from AI in the enterprise Consumer Internet Companies like Google are on the cutting edge and look for big breakthroughs. Sometimes this is not easy for people starting out. I would suggest to understand what your strengths are.
From a process perspective, I believe Depth is more important than Breath and having learned some Algorithms help but demonstrating value is more important and that is where most AI enthusiasts struggle
Follow Up — There’s so many topics and algorithms.. How do you get over analysis paralysis and pick a direction to develop depth in?
VK: What is the ROI of this experiment needs to be answered. Pick the one where your strengths lie and go from there.
QQ: I’m curious about your thoughts on what the entry level AI requirements are from a skills perspective?
Vimarsh Karbhari (VK):
- Basic Python
- Usage for Python Libraries
- Probability and Stats
These are foundational for getting into AI. I have two articles on my Blog which talk about these and some other important topics at length. I have aimed to provide some great links which are free to learn about these topics.
QQ: Follow up question — how would you market those skills to cater to an AI job? Happy to be directed to a link if that answers the question better.
VK: Learn by example is what I would suggest Stitchfix Algorithms tour. Basically build a portfolio of visualizations that demonstrate your skills.
QQ: In your opinion, which projects should be in a Github profile to cover AI field? On which topics?
There are all sorts of interesting things on github. All the Udacity projects are certainly something that can be showcased. Always focus on your strengths and put your best work out there. For inspiration check these articles.
Also I would suggest checking out things like turicreate — Turi Create simplifies the development of custom machine learning models. Using these creatively will also make you standout.
QQ: How do I change careers within my company without putting myself in a vulnerable spot to lose my current position?
VK: I would suggest to start with having conversations with people in the domain in which you want to switch. See what their expectations are and whether you can be set up for success in that role. If you feel it is something you can do then take the next step and approach them with a proposal. Always mention the ways in which you can provide value. It is very important. People become receptive when you can show them you can provide value. You always have to provide value first before you can get any value out of it.
QQ:What recommendations do you have for folks to practice for technical interviews?
VK: My blog lists all the questions of the companies which have open AI/DS/ML positions:
QQ: What advice do you have for people trying to decide between working with Startups, or more established companies?
VK: This has an interesting take. Data is the starting point to have any AI. By the virtue of this, if startups have very little data it is very difficult to have any algorithms on top of it. So if you want to be on the bleeding edge of AI the best thing would be to work at a company like Google which has massive amounts of data. There are startups who are working on data products specifically which are interesting. But then it’s important to see if what they say is AI is it actually AI or a simple inference algorithm.
QQ: Question about ageism in tech: “I’m over 30. I think it’s too late for me and I don’t know if I can catch up with the learning. What advice do you have for me?”
VK: It is never late. It’s only late if you think it is. Start small and find your interests. The tech field is too big to not accommodate folks if they have talent. And again as I said if you can demonstrate value, people will come to you.
QQ: A lot of students are worried about “future proofing” their career. What advice do you have for them around things to study, learn, create, etc.?
VK: The future is AI/DS/ML. Every software engineer will need to be a data engineer in the future. So for future proofing I would suggest having some data related courses from Udacity or at your University. The people who will know how to work with data will be the most sought after in tech.
This concludes the amazing AMA I did at Udacity.
Acing AI’s AI Interview Questions articles for Microsoft, Netflix, Google, Apple, Amazon, Facebook and Uber have been very engaging to the readers. As a followup, next couple of articles were on how to prepare for these interviews split into two parts, Part 1 and Part 2. To learn from experts please visit Acing AI expert Insights.
Subscribe to our Acing AI newsletter, I promise not to spam and its FREE!
Thanks for reading! 😊 If you enjoyed it, test how many times can you hit 👏 in 5 seconds. It’s great cardio for your fingers AND will help other people see the story.
