The State of MLOps 2025 Survey 🔥 That time of the year has arrived!! The survey to map out the State of Production MLOps in 2025 is now OPEN FOR RESPONSES 🚀🚀🚀 If you have a few minutes, your contribution will make a significant difference to the whole production ML ecosystem 🥳 The results will be shared as open source like last year!! You can add your response directly at: https://bit.ly/state-of-ml-2025 🔥 |
|
|
|---|
|
Stanford Study on AI Automation Impact Sharing an interesting new study from Stanford that has been making the rounds online recently modelling the relationship between AI and the impact in reduction of headcount, showing direct impact in roles that are in risk of AI automation (e.g. customer service, marketing, sales, software engineering, etc) - they present the following six conclusions: 1) Substantial declines in employment for early-career workers in occupations most exposed to AI; 2) Economy-wide employment continues to grow, but employment growth for young workers has been stagnant; 3) Entry-level employment has declined in applications of AI that automate work; 4) These employment declines remain after conditioning on firm-time effects, with a 13% relative employment decline for young workers in the most exposed occupations; 5) These labor market adjustments are more visible in employment than in compensation; 6) These patterns hold in occupations unaffected by remote work and across various alternative sample constructions. One consideration that comes to mind: I had seen similar charts before being challenged due to the close correlation between the rise/drop of COVID+zero interest rates which better explain the rise and drop of jobs. |
|
|
|---|
|
Python: The Documentary Python powers modern AI/ML in production, as well as a large percentage of the web; this is a great documentary on the beginnings, challenges and high-speed growth of the Python progarmming language. This was quite a nice documentary to watch, as it traces the origins of Python to a language called "ABC" which apparently is where some initial design paradigms were adopted by Guido. It is interesting to hear some of the key milestones that enabled the growth of Python, such as becoming open-source early on, as well as focusing on becoming an alternative to Perl. Python managed to get some pretty impressive wins with web platforms adoption (eg Dropbox, Instagram) as well as strong adoption in the scientific stack (eg Numeric->NumPy/SciPy). Exciting to see what the next decade of Python will hold, there are some exciting developments (e.g. removal of GIL, low-level interoperability, agentic stacks, typing, etc) |
|
|
|---|
|
A Software Eng Agent in 100 Lines Have you ever thought of building your own agentic software stack from scratch? Here is a 100-line implementation of a software agent in Python: In this implementation actions run as stateless processes and the message history is fully linear, so trajectories double as clean training data for FT/RL. Despite being such a tiny implementation it seems to achieve ~68% score on SWE-bench which is quite impressive, and it works with any LLM without tool-calling APIs, and is relatively simple to scale. The example comes with a CLI and Visual UI, and supports batch inference, a trajectory browser, and Python bindings, so it provides quite an end-to-end deep dive across all relevant contexts to get you started with your own agentic software engineer. |
|
|
|---|
|
Developer Productivity Metrics Reading List Measuring developer (and AI) productivity has now become a competitive advantage, and as a technical leader it's great to see the growing interest, and this is a great reading list from DX's Laura Tacho. Some of the resources include the latest framework from DX, including their AI productivity measurement framework, as well as the usual suspects such as SPACE, DORA, DevEx, etc. For production ML, it is also important to map these into MLOps: things like time-to-model deployment, change failure rate for ML performance metrics, and mean-time-to-resolve for models/data pipelines, and more nuanced topics like data quality. |
|
|---|
|
SpaCy Against LLM Maximalism The SpaCy team put together a while back a case against LLM maximalism, making a point that sometimes bringing an LLM for a simple NLP problem that involves a combination of speed / accuracy / simplicity, bringing LLMs to solve the problem may backfire. We are aware that end-to-end prompting can produce slow, costly, brittle, non-modular systems if not done leveraging best practices. Particularly for non-generative tasks like sentiment analysis, NER, and POS-tagging, etc can be tackled with traditional approaches with better latency, accuracy, and reliability (+ often speed). It still holds that in these use-cases LLMs can be used for rapid prototyping, as well as for even creating synthetic data from existing labels. Irrespective of the content, evaluation still remains key, no matter whether dealing with traditional NLP or LLM based development - and the SpaCy library is still the best out there. |
|
|
|---|
|
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. Upcoming conferences where we're speaking:
Other upcoming MLOps conferences in 2025:
In case you missed our talks:
|
|
|---|
| | |
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!
|
|
|---|
| | |
| | | | The Institute for Ethical AI & Machine Learning is a European research centre that carries out world-class research into responsible machine learning. | | | | |
|
|
|---|
|
|
This email was sent to You received this email because you are registered with The Institute for Ethical AI & Machine Learning's newsletter "The Machine Learning Engineer"
|
| | | | |
|
|
|---|
|
© 2023 The Institute for Ethical AI & Machine Learning |
|
|---|
|
|
|
The Institute for Ethical AI & Machine Learning
