Graham M.’s Post

Breakfast with Data Nerds at Fido Tuesday Morning, August 13th 8:30am - 10am Join the conversation! https://lnkd.in/gS6rAbpN We always start with introductions ... and we'll take some time to discuss: Projects - What are you working on that you'd like to share with the group? Jobs Opportunities - Recruiters/Businesses - What are you hearing about available roles now ... and upcoming? Job Seekers - Who's looking? For what roles? With what skills? Advice? Resources to share - Meetups/Best Reads/Courses to Consider/How (and what) are you learning? A Day in the Life of Data Professionals (Analytics/Engineering/Science/AI) - What's New? What's Critical/Essential? What's not working? What's top of mind? Come with your Questions & Answers!

Data without comment

I recently had a fascinating conversation with Russ B., Senior Director, Data and Software Engineering at Thrasio. Russ shared why it's always better to ask questions than make criticisms, how he overcame a case of imposter syndrome and offered some great advice to job seekers on how to take a data analytics approach to your job search. Check out the video below 👇 #DataAnalytics #JobSearchTips #ImposterSyndrome

The temptation for easy data teams is to add. What will let us fix this tech debt? Add more engineers What will allow us to build reports/dashboards faster? Add more tools How can we do the status quo and do new work? Add more contractors But it rarely works that easily: 🔵 You really needed the engineers 6 months ago to make an impact today 🔵 Tools don’t fix bad process 🔵 When you are relying on hero ball, every additional person adds less and less marginal capacity That is how you end up with data teams that feel like they are drowning and still be able to cut 1/4 of the head count with no impact to results. As data teams mature the processes have to mature with them. It is hard work, but there are no short cuts.

During a call yesterday, a director of data shared how, at his previous company (a well-known consumer fitness brand), they built their entire data platform for less than $100k a year with a team of four data engineers! This is impressive for a company that has millions of customers and lots of scale. He mentioned how it was a disciplined and conscious effort that took a few years of tuning and being ruthless about what was needed and what wasn’t. In contrast, we speak with teams that manage basic reporting and spend over $200k a year on tech with teams growing. For a long time, the modern data stack solution was to add: Build another model to understand x. Buy another tool to solve for y. Hire another engineer to build z. This isn’t a call to make teams smaller or to just cut for the sake of cutting down, but analytics doesn’t need to be expensive.

📣 Welcome to the Team, Daniel! 🎉 We are thrilled to introduce our newest team member, Daniel, who joins us as Data Scientist. Want to learn more about Daniel? Check out our exclusive interview where he shares insights about his background, passion, and what he looks forward to achieving with us. 🌟 👉 Read the whole interview by clicking the link. #MeetTheTeam #wearecraftworks #Team

Lets read Data

I'm really stoked with the progress of an ML model for a client. Here's what we've done: Ingested 20 years of history Run a massive decision tree based ML model Make a binary prediction Convert the prediction from binary to 0-100 Review the accuracy of the 0-100 scores to "what we got right" We're now at 28% accuracy for scores 90-100, which is phenomenal because the other 72% is basically their new lead list for business development. And, we can filter that 72% further if needed. The model is functional, based on real data, will soon be in their dashboards, and is really elevating their entire experience. Really proud of the team's work on this and what we've been able to deliver. P.S. Sometimes 28% accuracy is phenomenal. Other times, 99% is just barely passing. Just need to narrow in on a use case and figure out what is required for your situation.

Hi Lava Kafle, Financial markets are deeply complex and constantly evolving, offering a unique opportunity to explore the intricate dynamics that shape trading decisions. In this competition, hosted by Jane Street, you'll build a forecasting model using real-world data derived from production systems, which offers a glimpse into the daily challenges of successful trading. Total Prizes: $120,000 Entry Deadline: January 6, 2024 This challenge highlights the difficulties in modeling financial markets, including fat-tailed distributions, non-stationary time series, and sudden shifts in market behavior. It's a chance to engage with problems similar to those faced by Jane Street's quantitative teams, who have spent decades innovating in this field. Good luck, Ryan Holbrook Kaggle Data Scientist Kaggle, Inc 1600 Amphitheatre Pkwy Mountain View, CA 94043 This email was sent to [email protected] because you indicated that you'd like to receive news and updates about Kaggle. If you don't want to receive these emails in the future, please unsubscribe here. You can also change your preferences on your account's profile page by logging in at kaggle.com.

25 years in fracking, and you're still stuck doing data entry? Shift your focus to high-level decisions, not data entry. Let technology do the heavy lifting while you lead the way. If you haven't watched the first episode of Nerdin' Out with Cold Bore yet, what are you waiting for? Throw it on the computer while you're at work and catch up on all the data-driven insights you don't want to miss! Listen now on YouTube, Spotify and Apple 🎧📲 #OilAndGasTech #Completions #Data #FrackingInsights