└── README.md /README.md: -------------------------------------------------------------------------------- 1 | # Roadmap 2 | ### Note: this roadmap is my point of view for courses recommendations to be an AI Engineer. If one of the courses didn't fit you, you can find another one that fits you. The most important thing is to finish the main points of the roadmap regardless of the course you will study from. 3 | Before we have a talk about roadmap of how to be a good AI engineer, we must first answer the following question??? 4 | - What Does an AI Engineer Do? 5 | ## Roadmap: Roles and Responsibilities of an AI Engineer 6 | - Building AI models from scratch and assisting product managers and stakeholders with analysis and implementation. 7 | - Developing and maintaining AI infrastructures and product development. 8 | - Performing statistical analysis and interpreting the results to guide and optimize the company’s decision-making. 9 | - Automating the infrastructure used by the data science team. 10 | - Transforming machine learning models into APIs that can be integrated with other applications. 11 | - Training and retraining systems when necessary. 12 | - Running AI and machine learning experiments and tests. 13 | - Collaborating and coordinating tasks with other teams to promote AI adoption and best practices. 14 | ## Roadmap: What Skills Should an AI Engineer Have? 15 | - Learn programming languages - Proficiency in object-oriented programming languages like Python, C#, or C++ is needed to become an AI engineer. 16 | - An understanding of frameworks like Keras and Tensorflow to build AI solutions 17 | - Ability to build deep learning algorithms using neural networks — familiarity with CNNs and RNNs 18 | - Familiarity with cloud platforms like Google Cloud, Amazon AWS, or Microsoft Azure — you should be able to deploy and scale models using these platforms. 19 | - An understanding of software development methodologies like Agile or Scrum 20 | - Experience with cloud computing 21 | - Knowledge of Data science, data wrangling, and big data. 22 | ## Roadmap: Basic Computer Science Skills 23 | Consider obtaining a degree in a related field, as AI engineering requires a strong educational background. Typically, AI engineers hold a bachelor’s degree in computer science, data science, mathematics, or a related discipline. While it isn’t always required, a master’s degree in these AI-related fields can prove beneficial. 24 | - Typically, you must learn computer science basics. What is bits and bytes and how RAM and CPU works. The fundamental of computer science. 25 | ![roadmap](https://github.com/Mostafa-Samy-Atlam/Roadmap/assets/78164140/d1a77ef3-9ae0-47b8-be4f-936772a18ec1) 26 | ## Programming skills 27 | C++ & Data Structure and Algorithm 28 | You have to study all the following tutorials in this part. 29 | - __[C++ "Arabic Content"](https://www.youtube.com/playlist?list=PLCInYL3l2AajFAiw4s1U4QbGszcQ-rAb3)__ - This course would take around 14 days to complete (one hour a day). 30 | - __[OOP "Arabic Content"](https://www.youtube.com/watch?v=6U6WtWG3NrA&list=PL1DUmTEdeA6KLEvIO0NyrkT91BVle8BOU)__ - This course would take around 10 days to complete (one hour a day). 31 | - __[Data Structure and Algorithm "Arabic Content"](https://www.youtube.com/playlist?list=PLCInYL3l2AajqOUW_2SwjWeMwf4vL4RSp)__ - This course would take around 10 days to complete (one hour a day). 32 | 33 | Generally, you can decrease the overall duration of studying by increasing the number of studying hours per day. 34 | ## Python Programming Skills: 35 | You can choose one of the following courses: 36 | - __[Udemy Course](https://www.udemy.com/course/100-days-of-code/)__ - This is a very good course that explains all aspects of python (it would take a long time to study). 37 | + __YouTube Channel:__ 38 | If this was your choice, you have to study all the following four links: 39 | + __[Python Fundamentals](https://www.youtube.com/playlist?list=PLeo1K3hjS3uv5U-Lmlnucd7gqF-3ehIh0)__ 40 | - __[Pandas Tutorial](https://www.youtube.com/watch?v=CmorAWRsCAw&list=PLeo1K3hjS3uuASpe-1LjfG5f14Bnozjwy)__ 41 | + __[NumPy Tutorial](https://www.youtube.com/playlist?list=PLUcmakntVocWGSKXIsUn1J7Wm9ekpZ87G)__ 42 | + __[Matplotlib Tutorial](https://www.youtube.com/playlist?list=PLeo1K3hjS3uu4Lr8_kro2AqaO6CFYgKOl)__ 43 | - __[Python "Arabic content"](https://www.youtube.com/playlist?list=PL6-3IRz2XF5UM-FWfQeF1_YhMMa12Eg3s)__ 44 | 45 | This step would take around 3 weeks to complete. 46 | ## SQL (Structured Query Language) 47 | __[SQL "Arabic Content"](https://www.youtube.com/watch?v=Dj1zTZwbMOQ&list=PL85D9FC9DFD6B9484)__ 48 | 49 | This would take only one week to complete. 50 | ## Mathematics 51 | ### For mathematics, you can study crash courses that conclude the most important concepts in mathematics for AI. 52 | - __[Linear Algebra](https://www.youtube.com/watch?v=fNk_zzaMoSs&list=PLZHQObOWTQDPD3MizzM2xVFitgF8hE_ab&index=2)__ 53 | - This .__[video](https://www.youtube.com/watch?v=1VSZtNYMntM&t=3326s)__ is also a good video for summarizing most important math concepts for AI, it is around 100 minutes 54 | - __[Udemy crash course for AI mathematics](https://www.udemy.com/course/mathematical-foundation-for-machine-learning-and-ai/)__ 55 | 56 | All previous courses can be considered crash courses that would take only one week to finish. 57 | ### You can study the next courses after finishing the whole roadmap. 58 | - __[Mathematics for Machine Learning and Data Science Specialization "DeepLearning.AI Coursera"](https://www.coursera.org/specializations/mathematics-for-machine-learning-and-data-science)__ 59 | - This link contains 3 courses: 60 | - Linear Algebra for Machine Learning and Data Science. 61 | - Calculus for Machine Learning and Data Science. 62 | - Probability & Statistics for Machine Learning & Data Science. 63 | - __[Mathematics for Machine Learning Specialization "Imperial College London Coursera"](https://www.coursera.org/specializations/mathematics-machine-learning)__ 64 | - This link contains 3 courses: 65 | - Mathematics for Machine Learning: Linear Algebra. 66 | - Mathematics for Machine Learning: Multivariate Calculus. 67 | - Mathematics for Machine Learning: PCA. 68 | - __[MIT Probability Course](https://www.edx.org/learn/probability/massachusetts-institute-of-technology-probability-the-science-of-uncertainty-and-data)__ 69 | ## Machine Learning 70 | You can choose one of the following courses: 71 | - __[Machine Learning](https://www.youtube.com/playlist?list=PLeo1K3hjS3uvCeTYTeyfe0-rN5r8zn9rw)__ 72 | - __[Machine Learning "Arabic Content"](https://www.youtube.com/@HeshamAsem)__ 73 | 74 | This step would take around two weeks to complete according to your choice from these two courses, it would require 1 to 2 hours a day to finish in the specified duration. 75 | ## Deep Learning 76 | You can choose one of the following courses: 77 | - __[Deep Learning Udemy Course](https://www.udemy.com/course/deeplearning_x/)__ - This is a very good course that would take also around a month to complete. 78 | - __[DeepLearning.AI Specialization](https://www.coursera.org/specializations/deep-learning?utm_medium=sem&utm_source=gg&utm_campaign=B2C_NAMER_deep-learning_deeplearning-ai_FTCOF_specializations_pmax-nonNRL-within-14d&campaignid=20131140422&adgroupid=&device=c&keyword=&matchtype=&network=x&devicemodel=&adposition=&creativeid=&hide_mobile_promo&gclid=CjwKCAjwo9unBhBTEiwAipC112QQ1uqESaWhRpVNpqlYm8PS0s_FnfcvjOMnfDs0fUGbSf14YkeRThoCz_EQAvD_BwE)__ 79 | - __[Deep Learning "Arabic Content"](https://www.youtube.com/playlist?list=PL6-3IRz2XF5UiBoBDgeu5T3TyOIrgQ3r9)__ - If you choose this direction, it can take around a month to complete at maximum. 80 | - __[Deep Learning YouTube](https://www.youtube.com/playlist?list=PLZbbT5o_s2xq7LwI2y8_QtvuXZedL6tQU)__ - This is a small course that can be used as introductory course or for revision. 81 | ## Roadmap: Model Deployment 82 | ### Model Deployment: 83 | - After learning the topics mentioned above, you will can build AI applications from scratch. Now, you need to be able to deploy these applications and scale them. To do this, you need to learn how to put models in production with popular cloud platforms — Google Cloud, Amazon AWS, and Microsoft Azure. Pick any of these three platforms and start learning how to deploy models on them. Cloud computing is an essential skill for an AI engineer to have, so make sure to gain as much experience as possible in using these platforms. 84 | - This would take us to MLOPS, you should be good at one of the MLOPS frameworks. 85 | 86 | If you are interested in AWS platform, you can study one of the following 3 courses. All of these courses are about Amazon SageMaker which is a fully managed machine learning service that an help developers to build and train machine learning models, and then directly deploy them into a production-ready hosted environment easily and quickly. 87 | #### __Amazon-Sagemaker for machine learning:__ 88 | - __[Udemy Course for AWS-Sagemaker](https://www.udemy.com/course/practical-aws-sagemaker-6-real-world-case-studies/)__ 89 | - __[Coursera Course for AWS-Sagemaker](https://www.udemy.com/course/aws-machine-learning-a-complete-guide-with-python/)__ 90 | - __[Youtube Playlist for AWS Sagemaker "You can use this for revision"](https://www.youtube.com/watch?v=LkR3GNDB0HI&list=PLZoTAELRMXVONh5mHrXowH6-dgyWoC_Ew)__ 91 | 92 | If you are interested in Azure, you can study the following course. This course is about the tool provided by Azure for MLOPS. This tool has the same usage as the tool stated above (Amazon Sagemeker) except that it is a service provided by Azure - __[Azure tool for machine learning "Udemy Course for MLOps in Azure"](https://www.udemy.com/course/mlops-course/)__ 93 | 94 | ## Roadmap: Large Language Models: LLMS 95 | - Large Language Models (LLMs) are advanced artificial intelligence models trained on vast amounts of text data to understand and generate human-like language. 96 | - They are based on transformer architectures like GPT (Generative Pre-trained Transformer) and have a wide range of applications in natural language understanding, text generation, and conversation. LLMs, such as GPT-3.5, are among the largest and most powerful language models developed to date, capable of assisting with tasks like answering questions, providing explanations, and engaging in conversation with users. 97 | - __[YouTube Tutorial for LLMs](https://www.youtube.com/watch?v=MLLLDaR6P08&list=PLTPXxbhUt-YWSR8wtILixhZLF9qB_1yZm)__ 98 | --------------------------------------------------------------------------------