Issue date: 2nd September, 2024
Top AI and Cloud Skills
Lead Story
Dr. Vijay Srinivas Agneeswaran is the Senior Director and Principal Machine Learning Research Manager at Microsoft. He has extensive AI/ML/Data science and platform engineering experience. He can be reached at [email protected]
Top AI and Cloud Skills in 2024
Cloud computing and Artificial Intelligence are now essential for businesses. Companies no longer need to manage their own computing, storage, and networking infrastructure. AI is becoming pervasive, especially with the rise of Large Language Models (LLMs).
This article highlights the key skills students need to pursue careers in Cloud and AI.
There are five main skill categories, with some overlap:
- Data Science
- Data Engineering
- Machine Learning (ML) Engineering
- Full Stack Software Development
- AI Research
To stand out, it's important to show a solid understanding of cloud computing practices by earning certifications from major cloud vendors. Courses like Microsoft Certified: Azure Fundamentals are a good example.
Data Scientist
Though there's a distinction between data scientist, AI researcher, and ML engineer roles, the lines are often blurred in practice.
A typical data scientist might use ML algorithms to develop applications. ML engineers focus on deploying these models, while data scientists concentrate more on building them. This difference may disappear over time.
AI researchers, on the other hand, do more fundamental work. They focus on creating new algorithms, not just using them.
Students should prioritize programming skills and a strong foundation in probability, statistics, and machine learning.
Data Engineer
Students aiming for data engineering roles need a good understanding of database basics and some experience with NoSQL data stores.
Experience in building e-commerce, gaming, or IoT applications using NoSQL can be an advantage.
Working with open-source data processing tools like Apache Spark can also be beneficial.
Machine Learning Engineer
ML engineers typically build, evaluate, and deploy ML models as well as optimizing them using cloud technologies.
Both ML engineers and AI researchers need a strong theoretical background in statistics, probability, and machine learning. However, ML engineers can also benefit from a strong grasp of cloud computing.
Having cloud-specific ML engineering certifications, like Google's Professional ML Engineer course, can give you an edge.
Full-Stack Software Engineer
Full-stack software engineers develop applications that use AI and cloud technologies.
They are usually "generalists" who can quickly and efficiently learn new tech stacks, unlike "specialists," who may have deeper experience in specific areas but find it harder to switch.
A strong foundation in software engineering is necessary for aspiring full-stack engineers. Knowledge of AI fundamentals and cloud certifications are also beneficial. Experience with web development using REACT and NodeJS can be useful too.
AI Researcher
With all the buzz around ChatGPT and Gemini, it's natural for students to aspire for AI researcher roles. However, these roles can be more challenging than they seem.
Students need a deep understanding of statistics, probability, and machine learning basics. AI researchers focus on making LLMs learn with minimal training data and reducing the energy and cost of training and inferencing.
Energy efficiency is becoming a major concern for startups and large enterprises. Doctoral students who explore these areas may have an advantage.
Another important research area is Responsible AI, which includes transparency, accountability, and fairness.
Students who have worked with transformers or state-space models may also stand out. This is true, especially if they have built these foundational models or even fine-tuned these models. Students who have just used the LLMs as black-box models may not gain any specific edge over others.
Publishing in top conferences like CVPR, NeurIPS, ICML, AAAI, and ICLR can be helpful for landing roles in leading research labs.
While a PhD isn't necessary to pursue an AI research career, publishing in top conferences is often required. PhD students with publications in these conferences have an edge, but good internships can also open doors for beginner roles.
Quote of the Week
Andrew Ng: "AI is the new electricity. Just as electricity transformed every major industry 100 years ago, AI will also transform every industry."
Important Updates
[Ongoing] L2Pro Platform - Qualcomm and National Law University, Delhi have collaborated with CIPAM, DPIIT to provide a broad range of courses on Intellectual Property Rights: (Free registration)
[Ongoing] Turnip Cloud Computing Course - Completely hands-on working level understanding of Google Cloud: (Free registration)
[Ongoing] Big Data Analytics with Google BigQuery - Learn petabyte scale data analytics with this hands-on course: (Free registration)
Enjoyed the newsletter? Please forward to a friend or colleague. It only takes 10 seconds! Creating this one took 2 days!
New to The Innovation Post Newsletter? Signup