Stay up to date with what's going on in the community by subscribing to our public cal.
Coffee Session
MLOps as Tool to Shape Team and Culture
In the last year or so Ciro Greco, VP of AI at Coveo, had to approach MLOps from the perspective of an organization that needs to grow efficiently. The ML team at Coveo more than doubled in a few months, raising several issues about productivity, retaining talent, and honestly even just making sense of the org chart.
These are all happy problems because they’re
growing, but problems nonetheless. They all have to do with a better developer experience and stronger experimentation culture.
Also, some of these concepts might seem obvious but - believe Ciro - they often need to be evangelized a lot. Like a lot!
For a company that has essential ML components in the product (like Coveo which sells search and recommendation APIs) the ultimate goal is to have a bunch of ML capabilities embedded in the product.
To do that your ML engineers need to be trained in the business problem as much as they are in hyperparameter optimization. So having 20 people that do ML as a separate unit in your company might not be the right way to go.
On the other hand, if the business units are not prepared to absorb the ML engineers with the right infrastructure, tooling, and processes they will not make an impact.
Here is where MLOps became an active area of interest. Ciro works with Jacopo Tagliabue and together
they tried to approach this problem also from an MLOps and developer experience perspective.
Traversing the Data Maturity Spectrum: A Startup Perspective
We talked with Mark Freeman, a senior data scientist at Humu where he builds data tools that drive behavior change to make work better. His core responsibilities center around 1) building data products that reach Humu's end users, 2) providing product analytics for the product team, and 3) building data infrastructure and driving data maturity.
Key takeaways:
Data maturity is a spectrum that takes time to traverse and for companies to ultimately be capable of production ML.
Your organizations overall strategy is what determines the speed in which you traverse the data maturity spectrum.
Along the data maturity journey a data team has to make various long-term decisions... ensure such decisions don't block being able to implement ML in the future.
If you want lo learn more about Mark has to say jump in!
We’ve all been there. A model encounters an issue in production. What follows is often painful and time-consuming – a wash-rinse-repeat process of queries, calculating metrics, and writing
ad-hoc scripts to slice the data to figure out why. In all, over 84% of ML teams cite time-to-resolution as a pain point today.
In a new white paper, Arize offers a modernization blueprint for teams looking to solve problems faster. The paper covers:
🔎Best practices for ML performance troubleshooting
📙Definitions of ML performance tracing, data slice, and the math behind performance impact score
✂️A step-by-step guide to breaking down performance by slices and effective root cause analysis
Watch the whole thing here. Listen here or here. And if listening to podcasts aren't your thing or you don't have an hour to spare, Coffee Sessions Takeaways are for you.
We have an official MLOps community jobs board now. Post a job here and join our collective to get notified about all the members in the community looking for a job.
Looking for a job? Fill out your profile and let companies start fighting over you cause let's face it, you are in high demand right now!
Best of Slack is its own newsletter now. Sign up for it here.
See you in Slack, Youtube, and podcast land. Oh yeah, and we are also on Twitter if you like chirping birds.