.svg)
The fascinating journey of machine learning – from its beginnings to today's innovative applications and specialised professions.
Welcome to part 1 of our short and new AI series on specific IT roles and their high demand with a focus on artificial intelligence, machine learning and data science. Read about specialisation in the field of artificial intelligence and about ethics and security in AI in the following articles. This article is about the evolution of machine learning and the changing requirements and profiles.
Imagine that your coffee machine could not only prepare your favourite coffee but also predict when you need it most. Ten years ago, this might have sounded like science fiction. But we are actually on the threshold of an era in which such scenarios will become reality.
A brief definition of machine learning and artificial intelligence
Klären wir kurz die Basics, denn Machine Learning (ML) und künstliche Intelligenz (KI) sind zwei Begriffe, die oft in einem Atemzug genannt werden. Doch was steckt genau dahinter?
KI ist ein weites Feld der Informatik, das sich mit der Erschaffung intelligenter Maschinen befasst. Diese Maschinen sollen menschenähnliche Aufgaben ausführen können. ML hingegen ist ein Teilbereich der KI. Es konzentriert sich darauf, Computersysteme zu entwickeln, die aus Erfahrungen lernen und sich verbessern können, ohne explizit programmiert zu werden.
One definition of machine learning would therefore be: Teaching a computer how to perform a task without programming it.
https://www.youtube.com/watch?v=PeMlggyqz0Y (Machine Learning explained in 100 seconds, YouTube)
Now that we have roughly defined this field, we will next go back to the beginnings of ML.
The Evolution of Machine Learning
Machine learning has developed from its theoretical beginnings into a key technology that we all use every day. Let's take a brief excursion and journey through time from the ‘big bang’ to the present day.
The origins of machine learning go back to the 1940s and 1950s, when the first work on artificial intelligence and neural networks was carried out. In his 1950 essay ‘Computing Machinery and Intelligence’, Alan Turing introduced the famous Turing test: to test the intelligence of machines, a test is carried out to see if a human can distinguish one computer from another in a text-based conversation. Around the same time, Donald Hebb proposed a learning rule for neural networks that laid the foundation for future models.
The first machine learning algorithms were developed in the 1950s to the 1970s. Frank Rosenblatt's perceptron (1957) was one of the first neural networks, used for binary classification. However, it was limited by its inability to solve non-linearly separable problems. During this time, simpler algorithms such as k-nearest neighbours (KNN) and decision trees emerged, while researchers explored rule-based learning methods.
In the 1980s and 1990s, machine learning began to shift towards statistical learning and probabilistic models. Algorithms such as support vector machines (SVMs) and Bayesian networks were introduced, enabling better handling of uncertainties and more accurate predictions. At the same time, backpropagation was rediscovered for training neural networks, ensuring the basis for the neural networks that would dominate decades later.
The 2000s marked the era of big data and the rise of powerful ensemble methods such as random forests and gradient boosting. With the increasing availability of large data sets and improved computational tools, machine learning became more feasible for real-world applications such as recommendation systems and fraud detection.
The 2010s saw the revolution of machine learning, driven by advances in neural networks, particularly convolutional neural networks (CNNs) for image recognition and recurrent neural networks (RNNs) for sequence data. Breakthroughs like AlexNet in 2012 and Google's AlphaGo in 2016 highlighted the potential of deep learning. Towards the end of the decade, transformer models such as BERT and GPT revolutionised natural language processing.
Of course, machine learning continues to evolve – today with continuous advances in self-supervised learning, federal learning and ethical AI. The limits that AI systems can achieve are constantly being pushed. Accordingly, approaches and priorities for machine learning positions are shifting.
Machine learning positions from 2010 onwards
As described earlier, machine learning experienced rapid growth starting in 2010, driven by advances in deep learning, big data, and increased computing power. As demand for intelligent systems increased, specialised roles emerged within machine learning to address the unique challenges of building, deploying, and scaling ML models.
2010-2015: The rise of the data scientist
In the early 2010s, the profession of data scientist emerged and quickly became one of the most sought-after job titles in tech. Data scientists combined skills in statistics, machine learning, coding, and business acumen to extract insights from data. They were responsible for everything from data preprocessing to model building and evaluation, and communicating results to stakeholders.
This era coincided with the big data boom – this is when companies started collecting huge amounts of data – but they didn't have the expertise to turn it into actionable insights. Data scientists were expected to work with machine learning algorithms, focusing primarily on classical methods such as decision trees, random forests and early applications of neural networks.
2015–2020: Deep learning and specialised roles
When deep learning gained momentum around 2015 – thanks to the success of models like AlexNet and the availability of large data sets like ImageNet – the machine learning landscape began to transform. The complexity of deep learning models and the need to process large amounts of unstructured data like images or text led to the emergence of specialised roles:
Machine Learning Engineer: While data scientists focused on research and prototyping, machine learning engineers were responsible for deploying and maintaining machine learning models in production environments. This role requires advanced software engineering skills and the ability to integrate models into scalable systems.
Deep Learning Engineer: With the advent of convolutional neural networks for image processing and recurrent neural networks (RNNs) for sequential data, deep learning engineers emerged as specialists in designing and optimising complex neural network architectures using frameworks such as TensorFlow and PyTorch.
Data Engineer: As machine learning models became more and more data-intensive, data engineers were essential for building the infrastructure needed to handle and process large amounts of data. They focused on creating data pipelines, managing databases and ensuring data quality.
2020-present: The development of MLOps and further specialisation
As many organisations introduced ML models into their production, the need to manage the entire ML lifecycle led to the emergence of MLOps (Machine Learning Operations). MLOps brought together machine learning and DevOps practices and focused on automating the deployment, monitoring and updating of models in production. Alongside this, other new roles and specialisations are emerging to meet the growing demand in the fields of AI, data science and machine learning. The following are some of the emerging machine learning roles and positions that are becoming more relevant at both tech giants and emerging startups:
1. Machine Learning Engineer (MLE)
Job description: ML engineers are responsible for developing and implementing machine learning models and systems. They work closely with data scientists to bring prototypes and research models into production.
Skills: Strong programming skills (Python, Java, etc.), knowledge of ML frameworks (TensorFlow, PyTorch, etc.), understanding of software engineering best practices, and experience with cloud platforms.
Why it's trending: As more and more companies integrate machine learning into their products, the demand for engineers who can deploy and maintain these models is skyrocketing.
2. AI/ML Product Manager
Job description: AI/ML Product Managers focus on developing products that utilise machine learning models. They bridge the gap between technical teams, stakeholders and customers, ensuring that the product aligns with both business goals and technical feasibility.
Skills: A blend of business acumen and technical understanding of machine learning, as well as excellent communication and project management skills.
Why it's trending: As AI products become mainstream, product managers who understand the nuances of machine learning are needed to create great AI products.
3. Machine Learning Ops Engineer (MLOps)
Job description: MLOps specialise in the infrastructure, tools and processes needed to deploy, monitor and maintain machine learning models in production. They ensure that models are scalable, reliable and easy to update.
Skills: Knowledge of cloud platforms (AWS, Azure, GCP), CI/CD pipelines, Docker, Kubernetes and monitoring tools. Familiarity with ML frameworks is also crucial.
Why it's trending: As more and more models move from research to production, the need for robust ML DevOps pipelines is increasing.
4. Data Scientist (Specialising in ML)
Job description: While the traditional role of a data scientist is broad, many data scientists today specialise in machine learning and focus on developing, training and optimising models for specific business applications.
Skills: Strong skills in statistics, data processing and machine learning algorithms. Knowledge of Python, R and common ML libraries such as scikit-learn.
Why it's trending: As companies become more data-driven, they need experts who can apply ML to specific business problems and go beyond basic analysis.
5. Deep Learning Engineer
Job description: Deep learning engineers focus specifically on developing and fine-tuning deep neural networks (e.g. CNNs, RNNs, transformers). They often work on tasks such as natural language processing (NLP), computer vision and speech recognition.
Skills: Expertise in deep learning frameworks such as TensorFlow and PyTorch, a solid mathematical foundation and experience with GPU computing.
Why it's trending: Deep learning is at the heart of many AI breakthroughs, and therefore there is an increased demand for engineers who can push the boundaries of what deep learning models can do.
6. AI research scientist
Job description: AI researchers focus on exploring new areas of machine learning and artificial intelligence, and often work on cutting-edge algorithms, model architectures, and theoretical advances.
Skills: A strong academic background in mathematics, statistics and computer science. Knowledge of ML frameworks and an in-depth understanding of ML theory.
Why it's trending: Many companies are investing in AI research to stay competitive and explore new innovations that can provide breakthroughs in their respective fields.
7. Ethical AI Specialist
Job description: Ethical AI Specialists ensure that machine learning models and AI systems are designed to be fair, transparent, and free from bias. They help organisations address ethical challenges related to AI.
Skills: Knowledge of AI ethics, policy-making, techniques for reducing prejudice and legal frameworks related to AI. Familiarity with machine learning concepts is also essential.
Why it's trending: As AI systems become more widespread, ethical concerns around bias, fairness, and accountability are becoming more important.
8. AI Solutions Architect
JJob description: AI Solutions Architects design and implement AI-driven solutions tailored to specific business needs. They work closely with technical and non-technical teams to ensure the successful integration of AI into business processes.
Skills: In-depth knowledge of machine learning, cloud platforms and system architecture. Experience in translating business problems into AI solutions.
Why it's trending: As more and more organisations adopt AI, the demand for professionals who can design end-to-end AI systems is increasing.
9. Natural Language Processing (NLP) Engineer
Job description: NLP engineers specialise in developing models and systems that process and analyse human language. They work on tasks such as sentiment analysis, machine translation, text generation and speech recognition.
Skills: Knowledge of NLP frameworks such as Hugging Face, experience with deep learning, linguistic knowledge and familiarity with large language models (LLMs).
Why it's trending: With the rise of chatbots, voice assistants and language models like GPT, NLP is one of the most in-demand areas in machine learning.
10. AI Ethics Policy Advisor
Job description: This role focuses on advising governments and organisations on creating guidelines for the ethical use of AI. This may include drafting regulations, guiding the development of fair and transparent AI systems, and ensuring compliance with ethical standards.
Skills: Understanding of AI ethics, regulatory frameworks, legal knowledge and familiarity with machine learning systems.
Why it's trending: As governments and companies seek to avoid the misuse of AI, there is a need for professionals who can work at the intersection of technology, law, and ethics.
11. Robotics Engineer (with ML focus)
Job description: Robotics engineers specialising in machine learning design intelligent robots that can learn from their environment and make decisions independently. They work on integrating ML models into robotic systems to enable tasks such as object recognition, navigation and processing.
Skills: Experience in robotics, computer vision, control systems and machine learning. Knowledge of ROS (Robot Operating System) and ML frameworks.
Why it's trending: Robotics is a rapidly growing field. Integrating machine learning into robots is becoming a common approach to making them more autonomous.
12. AI hardware engineer
Job description: AI hardware engineers focus on developing specialised AI hardware (e.g. AI accelerators such as TPUs/Tensor Processing Unit, GPUs and company-specific chips) that are optimised to run machine learning algorithms.
Skills: Expertise in computer architecture, hardware design and parallel processing. Familiarity with machine learning algorithms is a distinct advantage.
Why it's trending: The computing requirements for training and executing deep learning models are growing, as is the demand for specialised hardware.
Conclusion
To conclude, let's summarise the key trends driving these roles:
- AI in production: As more companies move from research to production, roles are needed that focus on deploying, scaling, and maintaining ML models (e.g., MLOps Engineers, AI Solutions Architects).
- Ethics and bias: As AI systems become more ubiquitous, there is an increasing focus on ensuring fairness, transparency and accountability (e.g. Ethical AI Specialists, AI Ethics Policy Advisors).
- Specialisation: As machine learning matures, certain subfields such as NLP, computer vision and deep learning are becoming more specialised, leading to niche roles (e.g. deep learning engineers, NLP engineers).
These roles reflect the increasing sophistication of machine learning as it moves out of research labs and into the mainstream of modern technology products and services.
