Best Practices

What's wrong with technical interviews | Census

Nicole Mitich
Nicole Mitich August 03, 2022

Nicole Mitich is the content marketing manager @ Census. She's carried a love for reading and writing since childhood, but her particular focus is on streamlining technical communication through writing. She loves seeing (and helping) technical folks share their wisdom. San Diego, California, United States

Ah, the technical interview. YouTube has more than 100,000 videos for the interviewee, offering tips and tricks for acing the interview that assesses a data pro’s hard skills. But, based on the lack of cheat sheets for how to conduct these interviews, the hiring side must be pretty straightforward… Right? 

(Spoiler: It’s not. 😖)

In fact, because it’s so unclear, this was the topic of conversation for a recent episode of The Sequel Show. Census CEO and co-founder, Boris Jabes, talked with David Corea, Senior Analytics Manager at CoderPad, about why data leaders struggle to make sure technical interviews lead to good hires and what companies can do to fix them.

CoderPad offers a live coding platform to help gauge the technical abilities of job candidates in real-world scenarios (i.e. not by hand-writing code on a dry-erase board). According to David, CoderPad is largely driven by customer feedback – with interviewers saying, “I want to know what I’m doing right and what I’m doing wrong.” That way, interviewers can actually work on  improving the interview process instead of repeating their mistakes.

“The world is nowhere near as sophisticated as we think,” Boris said. “We’ve been hiring and interviewing and building the software industry for 40 years, and still, we don’t really even know if a question is a good question.” 🤔

So… What should the technical interview measure?

Years ago, software giants like Google and Microsoft famously asked job candidates to solve brain teasers, supposedly to test critical thinking skills. But eventually, companies realized that a person’s aptitude for riddles isn’t a good proxy for their ability to do a job.

“We need to know if the questions we ask are effective in assessing whether a candidate is up to snuff,” David said.

The technical interview should measure the hard skills the job requires, like writing SQL or JavaScript. To get a good sense of the candidate’s ability, interviewers should let them work with a familiar editor on unbiased problems that don’t require a deep knowledge of the company’s specific data.

The technical interview also presents the opportunity to find out if the candidate has the imagination and curiosity to put questions in context and learn the real goal.

“When someone asks an analyst a question, you don’t know if it’s urgent or if they’re just curious,” Boris said. “Or you might get a question and not realize it’s tied to a product decision about the online shopping cart that could shake the whole company. You have analysts who answer a question out of context without thinking about the repercussions.”

Improving the technical interview

The technical interview should give the interviewer confidence in the candidate’s ability to solve the problems their resumé says they can solve. Traditionally, that means the interviewer poses a hypothetical problem and asks the interviewee to solve it, staring at them while they try to rack their brain for the correct answer.

“Someone will write up something with a whiteboard marker and then they go back and forth, chatting over a whiteboard,” David said. “But that doesn’t actually capture what day-to-day work looks like for a developer. We need to make it easy for an interviewer to genuinely assess how a person’s hard skills will look in day-to-day work, in a way that makes it fun for the person going through the interview.”

Give job candidates a fair shot to show off their best work

Writing code on a whiteboard feels foreign (read: stressful) to interviewees and it doesn’t give interviewers a very accurate picture of their abilities. CoderPad closes the gap by allowing the interviewer to pick the tool they want to test (e.g., SQL). The platform then presents a bias-free SQL challenge the candidate can tackle then and there.

If interviewers really want to give candidates a fair shot, we need to keep interview tests as domain-neutral as possible. Realize that you and your team have spent months or years immersed in this domain and these databases; you’ll never find an outside candidate who can jump in and match your proficiency right off the bat. 

“I’m trying to see you at your best, not your worst,” Boris said. “If I can offer you the tools you’re comfortable with, you’re going to be more comfortable and better able to do your best work.”

While there’s no way to completely remove stress from the interview process (hello, there is a job on the line), companies can do a lot more to make the environment comfortable for candidates. Remember: Interviewing is a two-way street. 🛣️

“In every interview, there’s the candidate and someone who could potentially be a colleague,” David said. “The candidate is assessing, ‘Would I want to work with these people?’”

At Census, candidates for engineering roles end the interview process working with two of our full-time employees in a half-day hackathon. It’s a time investment, Boris said, but it gives us the sharpest picture of how a candidate could actually perform with our team if they get the job.

(Bonus: The candidate gets a sharper image of whether they would enjoy working here. Win-win. 😊)

Allow room to experiment

Looking for ways to improve technical interviews is like any honest scientific pursuit. When you first start exploring questions, you don’t know if they’ll lead nowhere or if they’ll give you actionable insight.

After years of asking the same interview questions, companies should be able to start benchmarking. I mean, if you can’t uncover a job performance trend related to the way candidates answer a routine question, you probably want to think about recalibrating the question or scrapping it altogether.

“The sad reality of companies is that as they grow, they start building rules and systems to allow for the average person to perform well rather than for the superstar to perform well,” Boris said. “You want to know if a question people succeed at is actually an indicator of future performance. Because the ultimate metric is: Are they a successful member of our team?”

How can you measure impact?

Every company in the world would love to measure what Boris calls the “super KPI” – otherwise known as the rate of learning.

“It’s almost like Evolution 101,” he said. “Learning is better than moving, because one implies the other, theoretically. How can we build a KPI for the rate of learning in an organization?”

Whether the goal is improving interview processes or business outcomes, David said, a focus on learning requires a collaborative effort to experiment and iterate on a macro level. Some of those experiments will yield actionable insights; some won’t. 🤷‍♀️

“If every insight you derive is actionable, you’re probably not trying hard enough,” Boris quipped. “If every swing is a hit, you’re not taking enough swings.”

When you’re evaluating improvements you can make to your technical interview process, some tweaks will be a home run, and some will teach you what not to do, but both are valuable to the interview process.

At the end of the day, improving technical interviews helps everybody. Companies can be more confident their hires can get the job done, and candidates have a more positive experience and enter the company on a high note.

Want to learn more? You can catch the full conversation between Boris and David below, or on your favorite streaming platforms. 🎧

Then head on over to The Operational Analytics Club to join the conversation around this and many other data best practices. ✨

Want to join the Census team? We’re hiring! ✨

Related articles

Product News
Sync data 100x faster on Snowflake with Census Live Syncs
Sync data 100x faster on Snowflake with Census Live Syncs

For years, working with high-quality data in real time was an elusive goal for data teams. Two hurdles blocked real-time data activation on Snowflake from becoming a reality: Lack of low-latency data flows and transformation pipelines The compute cost of running queries at high frequency in order to provide real-time insights Today, we’re solving both of those challenges by partnering with Snowflake to support our real-time Live Syncs, which can be 100 times faster and 100 times cheaper to operate than traditional Reverse ETL. You can create a Live Sync using any Snowflake table (including Dynamic Tables) as a source, and sync data to over 200 business tools within seconds. We’re proud to offer the fastest Reverse ETL platform on the planet, and the only one capable of real-time activation with Snowflake. 👉 Luke Ambrosetti discusses Live Sync architecture in-depth on Snowflake’s Medium blog here. Real-Time Composable CDP with Snowflake Developed alongside Snowflake’s product team, we’re excited to enable the fastest-ever data activation on Snowflake. Today marks a massive paradigm shift in how quickly companies can leverage their first-party data to stay ahead of their competition. In the past, businesses had to implement their real-time use cases outside their Data Cloud by building a separate fast path, through hosted custom infrastructure and event buses, or piles of if-this-then-that no-code hacks — all with painful limitations such as lack of scalability, data silos, and low adaptability. Census Live Syncs were born to tear down the latency barrier that previously prevented companies from centralizing these integrations with all of their others. Census Live Syncs and Snowflake now combine to offer real-time CDP capabilities without having to abandon the Data Cloud. This Composable CDP approach transforms the Data Cloud infrastructure that companies already have into an engine that drives business growth and revenue, delivering huge cost savings and data-driven decisions without complex engineering. Together we’re enabling marketing and business teams to interact with customers at the moment of intent, deliver the most personalized recommendations, and update AI models with the freshest insights. Doing the Math: 100x Faster and 100x Cheaper There are two primary ways to use Census Live Syncs — through Snowflake Dynamic Tables, or directly through Snowflake Streams. Near real time: Dynamic Tables have a target lag of minimum 1 minute (as of March 2024). Real time: Live Syncs can operate off a Snowflake Stream directly to achieve true real-time activation in single-digit seconds. Using a real-world example, one of our customers was looking for real-time activation to personalize in-app content immediately. They replaced their previous hourly process with Census Live Syncs, achieving an end-to-end latency of <1 minute. They observed that Live Syncs are 144 times cheaper and 150 times faster than their previous Reverse ETL process. It’s rare to offer customers multiple orders of magnitude of improvement as part of a product release, but we did the math. Continuous Syncs (traditional Reverse ETL) Census Live Syncs Improvement Cost 24 hours = 24 Snowflake credits. 24 * $2 * 30 = $1440/month ⅙ of a credit per day. ⅙ * $2 * 30 = $10/month 144x Speed Transformation hourly job + 15 minutes for ETL = 75 minutes on average 30 seconds on average 150x Cost The previous method of lowest latency Reverse ETL, called Continuous Syncs, required a Snowflake compute platform to be live 24/7 in order to continuously detect changes. This was expensive and also wasteful for datasets that don’t change often. Assuming that one Snowflake credit is on average $2, traditional Reverse ETL costs 24 credits * $2 * 30 days = $1440 per month. Using Snowflake’s Streams to detect changes offers a huge saving in credits to detect changes, just 1/6th of a single credit in equivalent cost, lowering the cost to $10 per month. Speed Real-time activation also requires ETL and transformation workflows to be low latency. In this example, our customer needed real-time activation of an event that occurs 10 times per day. First, we reduced their ETL processing time to 1 second with our HTTP Request source. On the activation side, Live Syncs activate data with subsecond latency. 1 second HTTP Live Sync + 1 minute Dynamic Table refresh + 1 second Census Snowflake Live Sync = 1 minute end-to-end latency. This process can be even faster when using Live Syncs with a Snowflake Stream. For this customer, using Census Live Syncs on Snowflake was 144x cheaper and 150x faster than their previous Reverse ETL process How Live Syncs work It’s easy to set up a real-time workflow with Snowflake as a source in three steps:

Best Practices
How Retail Brands Should Implement Real-Time Data Platforms To Drive Revenue
How Retail Brands Should Implement Real-Time Data Platforms To Drive Revenue

Remember when the days of "Dear [First Name]" emails felt like cutting-edge personalization?

Product News
Why Census Embedded?
Why Census Embedded?

Last November, we shipped a new product: Census Embedded. It's a massive expansion of our footprint in the world of data. As I'll lay out here, it's a natural evolution of our platform in service of our mission and it's poised to help a lot of people get access to more great quality data.