Issue #299 - The ML Engineer ??
Alejandro Saucedo
Tech Executive @ Zalando | Chair/Advisor @ UN, EU, ACM, etc | Join 60k+ ML Newsletter
Thank you for being part of over 60,000+ ML professionals and enthusiasts who receive weekly articles & tutorials on Machine Learning & MLOps ?? You can join the newsletter for free at https://ethical.institute/mle.html ?
If you like the content please support the newsletter by sharing with your friends via ?? Email, ?? Twitter , ?? Linkedin and ?? Facebook !
If you are a Machine Learning Practitioner looking for an interesting opportunity, I'm currently hiring for an Applied Science Manager for Forecasting & Causal Inference for a Senior Applied Scientist for Forecasting - do check it out and do feel free to share broadly!
This week in Machine Learning:
Canva's product analytics pipeline processes 25 billion events daily, with growing use-cases across A/B testing, personalization, and insights - this is a great deep dive into how they make it happen: Canva has adopted key principles to enable massive-scale processing, such as ensuring events following a strict schema (using Protobuf), and enforcing schema with their internal service Datumgen. They collect events through a unified client, which they then ensure are validated, enriched, and routed via AWS Kinesis.
Uber brings together their approach to supporting Tiered Storage in Kafka, which ensures decoupling between Kafka’s storage and compute resources, which enables scalable and cost-efficient data retention. Uber achieves this by introducing local storage for recent data and remote storage (e.g., S3, HDFS) for older data, which supports for optimisation ?of resource use and reduces operational complexity. This is quite a practical deep dive, and it is great to see the close collaboration with the open source community, with the core Kafka team also contributing closely to extend and support these use-cases.
Things I Wished More Developers Knew About Databases: Often fundamentals of database design are overlooked by developers when working with databases, which can have a substantial benefit to their day to day development. Some of the key lessons include: 1) the fallibility of network reliability, 2) the varied interpretations of ACID across databases, and 3) the trade-offs between consistency, isolation, and performance.
The Effects of Generative AI on High Skilled Work: Evidence from Three Field Experiments with Software Developers. A great review on the state of GenAI bringing together three field experiments that examining the impact of GitHub Copilot on software developers' productivity. This study was conducted at Microsoft, Accenture, and an anonymous company - the study evaluates the effect of Copilot on metrics such as pull requests, commits, and successful builds (indeed not optimal metrics but still relevant insights). The results show that Copilot increases developer productivity by approximately 26%, with a particularly strong effect on the number of pull requests and builds. Junior and short-tenure developers exhibited higher adoption rates and saw the most significant productivity gains. However, the effects were not consistently statistically significant across all metrics or companies - this is obviously key, but it's great to see investments towards quantifying the causal impact of AI tools in development.
One of the age-old lessons on clean-code: the importance of writing explicit, clear code over implicit, ambiguous code in software development. For machine learning practitioners, this is especially relevant when working in large, collaborative codebases, as maintainability and clarity are always important. Implicit code can introduce confusion and increase the cognitive load for developers, leading to higher “WTF per Minute” (WTFPM) rates - indeed a great standardised quantifiable metric across industry (!).
领英推荐
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 2024:
In case you missed our talks:
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.
? 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!
OSS: Policy & Guidelines
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:
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 !
About us
The Institute for Ethical AI & Machine Learning is a European research centre that carries out world-class research into responsible machine learning.
Tech Lead, Search & Recommendation systems
2 个月Hey Alejandro Saucedo, thanks for write up! However, I think the link to Canva's 25B Events per Day is redirecting to an LLM workshop post.