Subscribe to the Machine Learning Engineer Newsletter

Receive curated articles, tutorials and blog posts from experienced Machine Learning professionals.

Issue #259 🤖 
If you like the content please support the newsletter by sharing with your friends via ✉️ Email, 🐦 Twitter, 💼 Linkedin and 📕 Facebook! If you've come across this you can join the newsletter  for free at
The ML Engineer this week:
Thank you for being part of over 50,000+ ML professionals and enthusiasts who receive weekly articles & tutorials on Machine Learning & MLOps 🤖 You can join the newsletter for free at
If you are a Machine Learning Practitioner looking for an interesting opportunity, I'm currently hiring for a few roles including Applied Science Manager, Applied Scientist, Analytics Team Lead, and Customer Analyst - do check it out and do feel free to share broadly!
Instacart's Machine Learning Platform Griffin 2.0: A great overview on Instacart's advanced machine learning platform, extending to model management, and deployment automation. It showcases a unified web interface, supports distributed ML tasks, and is built on a Kubernetes platform for centralized training workloads. They introduce important features in the MLOps space including standardized ML runtimes, horizontal scalability using Ray, and a comprehensive metadata store for effective model lineage management. The platform streamlines the model development lifecycle, offering a balanced approach between flexibility and standardization, and integrates model serving and feature engineering for a holistic ML workflow. This evolution reflects interesting lessons learned in unifying ML training solutions and considering broader application contexts.
You can now enable large language models run your computer: An innovative project designed to enable multimodal models like GPT-4v to autonomously operate a computer using human-like mouse and keyboard actions. It provides some interesting results even though it currently faces limitations, particularly in accurately estimating mouse click locations. The project is open source, and with opportunities for the broader AI community to support the evolution of this interesting framework.
Statistical vs Deep Learning forecasting methods: In Nixtla's analysis comparing deep learning (DL) models with statistical ensembles for time series forecasting in the M3 competition, they showcased how a simple statistical ensemble could outperform most individual DL models in these contexts. This raised an interesting tradeoff that also showcased the advantages in efficiency and cost; namely running 25,000x faster at a cost of $0.5c, compared to over 14 days and approximately USD 11,000 for the DL ensemble. This suggests that in contexts where resources and simplicity are critical, statistical methods should be considered before moving to more complex and costly DL models. These outcomes are complementary to the growing opportunities arising in deep learning transformed based architetures for forecasting, supporting accurate forecasting at massive-scale; we are expceted to see an interesting development across both DL and Stat FC in the near term.
DeepMind has introduced a mindblowing music generation AI model. They have taken an interesting approach, partnering with musicians to enable creative music practitioners to generate music with next-generation AI technology. These developments show a glimpse of the disruption coming to the music industry, offering nuanced control over music generation with capabilities in instrumental and vocal compositions whilst allowing for AI-generated music in various artists' styles. This emphasises the approach of responsible AI, introducing concepts such as "SynthID" for watermarking AI-generated audio to protect artist integrity and promote creative expression - however there is yet to see the opportunities and challenges that this will bring.
Code is read more times than it is written, however it's run more times than it's read, and it's used even more than it is run - very insightful way of thinking about your code: We have all heard the phrase "Code is Run More Than Read", which tends to emphasise the importance of code readability and maintainability, which can be thought as "maintainer > author". However we have to consider the importance of user experience and operational aspects over traditional coding practices. We can do this by thinking of hierarchy of user > ops > dev, underscoring the significance of operational costs and challenges in production. Beyond B2C contexts, in B2B it is worth taking business considerations into this framework, which would give us "biz > user > ops > dev", highlighting the importance of organizational goals and constraints but also making sure we don't bump into common challenges this brings. These include: unusable software (dev > user), works only on my machine (dev > ops), the wrong thing (dev > biz), resume-driven development (dev > *), and a few others - heck out this article for the deeper dive.
Upcoming MLOps Events
The MLOps ecosystem continues to grow at break-neck speeds, making it ever harder for us as practitioners to stay up to date with relevant developments. A fantsatic way to keep on-top of relevant resources is through the great community and events that the MLOps and Production ML ecosystem offers. This is the reason why we have started curating a list of upcoming events in the space, which are outlined below.
Relevant upcoming MLOps conferences:
Open Source MLOps Tools
Check out the fast-growing ecosystem of production ML tools & frameworks at the github repository which has reached over 10,000 ⭐ github stars. We are currently looking for more libraries to add - if you know of any that are not listed, please let us know or feel free to add a PR. Four featured libraries in the GPU acceleration space are outlined below.
  • Kompute - Blazing fast, lightweight and mobile phone-enabled GPU compute framework optimized for advanced  data processing usecases.
  • CuPy - An implementation of NumPy-compatible multi-dimensional array on CUDA. CuPy consists of the core multi-dimensional array class, cupy.ndarray, and many functions on it.
  • Jax - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
  • CuDF - Built based on the Apache Arrow columnar memory format, cuDF is a GPU DataFrame library for loading, joining, aggregating, filtering, and otherwise manipulating data.
If you know of any open source and open community events that are not listed do give us a heads up so we can add them!
As AI systems become more prevalent in society, we face bigger and tougher societal challenges. We have seen a large number of resources that aim to takle these challenges in the form of AI Guidelines, Principles, Ethics Frameworks, etc, however there are so many resources it is hard to navigate. Because of this we started an Open Source initiative that aims to map the ecosystem to make it simpler to navigate. You can find multiple principles in the repo - some examples include the following:
  • MLSecOps Top 10 Vulnerabilities - This is an initiative that aims to further the field of machine learning security by identifying the top 10 most common vulnerabiliites in the machine learning lifecycle as well as best practices.
  • AI & Machine Learning 8 principles for Responsible ML - The Institute for Ethical AI & Machine Learning has put together 8 principles for responsible machine learning that are to be adopted by individuals and delivery teams designing, building and operating machine learning systems.
  • An Evaluation of Guidelines - The Ethics of Ethics; A research paper that analyses multiple Ethics principles.
  • ACM's Code of Ethics and Professional Conduct - This is the code of ethics that has been put together in 1992 by the Association for Computer Machinery and updated in 2018.
If you know of any guidelines that are not in the "Awesome AI Guidelines" list, please do give us a heads up or feel free to add a pull request!
© 2023 The Institute for Ethical AI & Machine Learning