News
Happy February! Here's all the news to keep you up to date.
We've hired and are still hiring!
We have four new team members this month!
Dave Berenbaum came to Iterative.ai by way of a previous contribution to our open source products while working as a Data Science Manager at Captial One. He joins the team as a Technical Product Manager. We are thrilled he's here!
Batuhan Taskaya joins us as a DVC Software Engineer working on the Python core. Batuhan is excited to work on open source full time and we are excited to have him do so!
Jeny De Figueiredo is involved in the Seattle area data science community at Data Circles and is a WiDS Puget Sound Ambassador. She joins us as our new Community Manager and is looking forward to further building and engaging the community in MLOps! (Hi! This is me. ππ»ββοΈ I'll be writing Heartbeat!)
Roger Parent has already been a big part of building DVC and CML. He has been a primary developer of a UI that interfaces with the DVC Python application to provide an interface with the Experiments feature that's coming out with DVC 2.0. We are so excited to have him joining us full time as Software Engineer.
Open Positions
We are on the hunt for a TypeScript Front-End Engineer to build SaaS and a VS Code UI for our popular machine learning tools: DVC and CML. The ML tools ecosystem is what JS space was 10 years ago. Come join us on this exciting project!
Our search continues for a Developer Advocate to support and inspire developers by creating new content like blogs, tutorials, and videos - plus lead outreach through meetups and conferences.
Does this sound like you or someone you know? Be in touch!
Iterative.ai Featured on The New Stack
Susan Hall of The New Stack.io interviewed our very own CEO, Dmitry Petrov, discussing the needs of ML engineers and how Iterative.ai makes tools to enable version control and CI/CD for versioning data and ML models.
"ML engineers, they still need collaboration. They need GitHub for collaboration, they need this CI/CD system to resolve [issues] between each other, between the team and productions system." - Dmitry Petrov
Learning Tools for ML Engineers
Workshops and Talks
Developer Advocacy for Data Science
So you saw the post further up. ππ½ Curious about developer advocacy or what to look for in a hire for this position? Elle O'Brien dove into this recently with Alexey Grigorev (author of a Data Science Bookcamp) in this podcast on DataTalks.club You can watch it here below. ππΌ
From the Community
As ever, we have much to share from the great citizens of the DVC community.
spaCy and DVC Integration
If your NLP team uses spaCy to manage your projects, with spaCy's release of v3.0, you can now enjoy DVC integration to manage your workflow like Git! Check out the documentation here to streamline and track your process! π
spaCy Integration
DagsHub and DVC Integrations
This month two great articles came out regarding the integration of DAGsHub and
DVC. First, this article: [Datasets Should Behave Like Git Repo walks you
through the steps to use DVC in your data versioning. The following image shows
the dependencies and how you simply need to do a dvc update
each time your
dataset or model changes to track the process.
Datasets Should Behave Like Git Repositories
Did you say "Works Out of the Box?"
Also from DAGsHub, by CEO Dean Pleban, Free Dataset & Model Hosting with Zero Configuration - Launching DAGsHub Storage tells how their new DAGsHub storage is a DVC remote that requires zero configuration (!) and will allow for team and organization access controls as well as easy visibility.
Model Management and ML Workflow Orchestration with DVC and Apache Airflow π©πͺ βοΈ
We're really excited about a German language workshop led by Matthias Niehoff! The workshop will be a part of the ML Summit 2021 taking place April 19-21st, but registration closes February 18th. So time is ticking. β° The Conference is online, but will be in German. For more info, head here ππ½ for the Workshop Details.
"The most popular 'N+1' tool used by teams on Spell"
Using DVC as a Lightweight Feature Store on Spell by Aleksey Bilogur , reviews the process of using DVC with Spell for managing changing datasets, enabling team-wide data reproducibility and why Spell fans are DVC fans, and vice versa. π
Tweet Love β€οΈ
How do you deploy a machine learning model?
β Mihail Eric (@mihail_eric) February 3, 2021
Check out my new post! https://t.co/Qx3RtQ7hO0
In it we build:
πREST service with @tiangolo's sleek FastAPI
πChrome extension to interact with the model
π³Custom @Docker images
πCI/CD with @DVCorg + Github actions
You're all caught up! See you at the next Community Gems π!
Do you have any use case questions or need support? Join us in Discord!
Head to the DVC Forum to discuss your ideas and best practices.