How to become an AI Engineer

Hello, Can Anyone Share the Roadmap to becoming an AI Engineer?

1 Like

Hi @CHAITANYA_DAVE , Welcome to the DeepLearning.AI platform. We apologise for the late reply.
I think what you’re looking for is ML engineer. Now the roadmap depends on what type of career you want to pursue in future. For going into research in ML and AI, you should work and lay more emphasis in terms of mathematical knowledge and statistics and theoretical machine learning along a bit of knowledge of implementation while in case of an ML engineer, you should be well versed with different upto date frameworks, libraries and end to end implementation of model along with a decent amount of mathematical knowledge. Although both roles have a lot of overlapping area but then too the amount and type of knowledge required is different.

3 Likes

Hi @CHAITANYA_DAVE,

Becoming an AI Engineer requires a combination of technical skills and knowledge in various fields such as computer science, mathematics, and statistics.

  1. Learn Programming Languages
  2. Mathematics and Statistics
  3. Machine Learning
  4. Learn Frameworks and Libraries
  5. Data Wrangling and Preparation
  6. Develop Real-World AI Projects
  7. Stay Up-to-date
  8. Soft Skills

These are skills that I know and I hope they are helpful.

1 Like

Hi Dave,

in this thread recently a similar question was asked.

In my opinion the best personal sequence for you depends on:

  • where you stand now (e.g beginner or medium)
  • what you want to achieve (e.g. become an AI engineer in the field of IoT / Automotive)
  • your strength and background of the industry you want to work (e.g. background in image processing with focus on sensor fusion and deep learning)
  • your timeline, considering how much time you want to invest in your learning roadmap (e.g. 5 hours per week for 8 months or so)

If you have more context or questions, just let us know, @CHAITANYA_DAVE.

Best regards
Christian

1 Like

As I see there is big specific to TensorFlow. I need ML for text mainly. So, can I not learn TensorFlow Developer Professional and start to learn Natural Language Processing (NLP) Specialization immidiately after Deep Learning? Can you describe how is TensorFlow important at all and why do I need it? Can I use any alternative frameworks maybe? As I see, enough much time takes the learning TensorFlow by itself.

Hi @someone555777 , Thanks for posting this query. The course offered by DeepLearning.AI uses TensorFlow as the framework for the NLP course and If you don’t have the time to do the TensorFlow Developer Professional Certificate you could very well jump into the course as you have likely done the Advanced Learning Algorithms course. You could very well cope up with the teaching and if you find yourself in a catch you could just refer the documentation that ought to clear your doubts or you could connect with me or the Discourse Community. I’ll be happy to help you anytime. On the flip side you mentioned that you’d do the course after Deep Learning if you intend on learning it from the Deep Learning Specialisation offered by DeepLearning.AI then it covers all the TensorFlow concepts parallelly for you to just jump into the NLP Specialisation. As for the framework choice TensorFlow has a very robust community so that is probably the case for the course choice.

As for why TensorFlow is Important at all is -

It is an important open-source software library for machine learning and artificial intelligence tasks because it provides a declarative approach to defining the structure of your algorithm as a tree of nodes, which can be executed in an efficient and parallelised manner. It also provides efficient implementations of the most critical parts of fitting and evaluating a complex model, as well as tools for visualising the operation and state of the model. Additionally, TensorFlow’s use of a predefined graph structure can help reduce coding errors and simplify implementation.

On the other hand you could use could use any alternative framework for NLP like Pytorch, Stanford CoreNLP etc. It comes down to your own application, what caters best for your needs and your personal preferences.

Hope this clears your doubts

Thank you For such important guidance