Category: Guest Blogs

Since I was a pasty, awkward middle schooler I’ve been writing computer code. Much of my coding time is spent fixing my own silly bugs, but a couple of magical coding moments standout:

• When I accessed my hideous, lets animate-all-the-gifs AOL-hosted homepage on a middle school computer.
• Character-by-character, copying Tetris onto my TI-85 calculator from a friend.
• Watching the How to build a blog engine in 15 minutes video with Rails video. Rails was the secret weapon for small teams with big dreams.

Whether it’s skepticism honed by time or just reality, I’ve experienced fewer of these of moments over the past ten years. That is, until recently.

About a year ago, I solved my first problem with machine learning, identifying handwritten digits with only a few lines of code. It was magical taking something that seemed so human and solving it with so little code. There’s no way I could solve this problem elegantly with my traditional toolset. The problem I worked on is only scratching the surface of what machine learning can do. For example, machine learning is used today to:

• Detect breast cancer
• Drive a car
• Detect emotion
• Make your Facebook newsfeed addictive

It’s possible machine learning and the larger field of artificial intelligence will be the largest technological leap since the Internet.

So, where to begin? And should you even try?

There’s clearly demand for jobs in machine learning. Notice the large gap in employer demand vs. job seekers:

• from a Cornell University study

A major reason for this gap is that many companies are mistakenly looking for machine learning researchers, not candidates that know how to apply machine learning. Paraphrasing Cassie Kozyrko, Google’s Chief Decision Intelligence Engineer: if you run a bakery, you don’t hire an oven maker. You hire someone that knows how to bake bread. There’s a small minority of ML problems that require a new algorithm. Most problems can be solved with off-the-shelf models.

This hiring mindset is changing, but the wheels of big business move slowly. Today is a great time to take advantage of this gap: it’s significantly easier and cheaper for small companies to collect large amounts of data than it was a decade ago. Many smaller software companies have large enough data sets to solve time-consuming, differentiating problems. However, very few small companies have an applied machine learning engineer on staff to take advantage of all that data.

If you’re like me (machine learning-curious and too lazy to go back to school), applied machine learning might be for you. At 11AM Tuesday the 26th of February at The Innosphere, I’ll be giving a talk on getting started with machine learning. You’ll learn enough to talk intelligently about machine learning at your next network event, see me live-code a machine learning problem, and receive a curated list of resources to get you started.

In short, take advantage of my wasted efforts, false starts, and wrong turns. Just get the good stuff for applied machine learning.

Whatever pulls you into the startup space, you’ll inevitably find yourself face-to-face with one of the toughest requirements for your venture’s success: cold, hard sales.

At Sunshower.io, we talk to a lot of people about their cloud infrastructure usage. In our professional lives, we’ve dealt with the confusion caused by different cloud vendors, including confounding billing methods, lack of insight into the infrastructure you’ve built, and just throwing hardware and money at the current problem and hoping it’ll fix it. Understandably, the question we’re most frequently asked is the one that’s most mission-critical: How did my cloud bill get like this and how do I get it down?

1) You Forgot About Some Infrastructure

“Cloud sprawl” is extremely common, and happens when you’re running more cloud instances than necessary. It’s easy to see how this can happen—running workloads that you’ve forgotten about and unused and idle workloads are all key culprits. In a complex cloud ecosystem, it can be tough to keep watch over everything running in the cloud. Monitoring and controlling those workloads is key to making sure you’re not over-spending on the cloud.  If your company isn’t using auto-scaling, you might be running instances 24/7 that aren’t always performing a necessary function. Running instances that you’re not using is essentially throwing money away—like going away for the weekend and leaving all of your lights on.

2) You Bought Too Much “Just In Case”

Overprovisioning refers to buying more cloud resources than you typically need. It’s important to tailor what you buy to actual usage, because it really adds up. The first step is figuring out what you’re actually using, which monitoring  and optimization tools can help with. If this process is overwhelming, there are vendors you can work with to help you sift through your options and make the best possible choices. Without good monitoring tools, it’s impossible to what you’re wasting. Only then should you start looking into what to buy instead.

3) You Drank The Vendor Kool-aid

The custom services provided by cloud service providers are tempting, but the cost can really add up. Even worse, it removes your ability to migrate to other cloud providers, so it’s hard to pivot to more cost-effective solutions over time. As you build your cloud strategy, try to avoid locking yourself into a relationship with a single cloud service provider. Don’t tie yourself to a single vendor because it’s convenient—make sure that you’re allowing yourself the flexibility to change providers and adapt new strategies when costs start to increase.

Setting Yourself Up For Future Success

When it comes to cloud costs as a whole, think about it this way: When you build a snowman, you start with a tiny ball. As you roll it around, it picks up more and more snow until the ball is eventually so big you can’t even move it. No way are you picking that guy up—he’s staying right where he is until the inevitable destruction by meltdown. Cloud costs can incrementally build up (and melt down) in much the same way. Not everyone has a full-time IT department or the expertise to be able to game the system and make sure their cloud infrastructure is as optimized as possible.

The good news is, there are tools out there to put you on the path to reducing your cloud costs today. The trick is choosing the right solutions—ones priced for the size of your company that simplify your life on the cloud, rather than complicate it. Choosing the right tools to help avoid sprawl, overprovisioning, overspending are vital parts of a company’s survival. Make it a priority to understand how you use the cloud today, and you’ll be in a better position to reduce cloud spending tomorrow.

About Lisa Gumerman & Sunshower.io

Lisa Gumerman is the CEO of Sunshower.io, which offers cloud management solutions with a lower barrier to entry, focusing on turn-key solutions that don’t require installation or a complicated configuration process. Whether it’s visualizing and tracking cloud infrastructure, deploying applications across clouds, or economizing and using space more efficiently, we simplify the complicated task of working on the cloud. Our initial product launch optimizes AWS EC2 spend by 40 to 80% and is free until June 2019.

You can see Lisa’s panel Managing Technical Debt on Monday February 25th, 12:30pm-1:30pm @ The Articulate.