Scede’s Scaling So Far Podcast with Jonathan Siddharth
Scaling So Far shares candid conversations with tech founders and leaders on how they’ve built and scaled their teams.
You’ll want to enjoy this entire podcast here.
In the 50th episode of the Scaling So Far podcast, Jonathan Siddharth, Founder, and CEO of Turing discuss the lessons he’s learned scaling the Turing team with Dan, the Scaling So Far host. Jon and Dan also talk about why unlocking global opportunities for top tech talent and tactics for sourcing, engaging, and nurturing brilliant engineers are critical today.
Jonathan, pleased to be chatting with you today. Firstly, thank you for joining us on the Scaling So Far podcast – great to have you with us. Could you tell us a bit about yourself to kick things off?
Thank you for having me, Dan – excited to be a part of the podcast. For the listeners out there, I’m Jonathan Siddharth, CEO and Co-Founder of Turing. Turing is a platform that lets you push a button to hire prevetted remote engineers worldwide.
Turing uses artificial intelligence to automatically source software developers from all over the world, automatically vet them, match them, and help manage the collaboration after we match software engineers with opportunities.
We recently became a Unicorn a little over three years after we launched, and we are in the process of rapidly scaling. We call this process “blitzscaling,” where you grow at this accelerated pace. So excited to share some of our lessons learned and challenges with you.
Can you tell me a bit more about Turing’s mission and vision?
Absolutely. So we now live in a remote-first world, and every company today is in a race to reap the benefits of remote engineering talent. Twitter is going remote; Square is going remote; Coinbase is going remote – even traditional companies like Siemens, Ford, etc., are seeing the benefits of going remote.
And the reasons are obvious.
Number one, you can tap into a planetary pool of engineers versus just looking in your backyard.
Number two, you have the opportunity to tap into geographies that nobody else is looking at today, like Latin America, Africa, Southeast Asia, Central Europe, etc.
And third, distributed teams work now, as we’ve discovered in the last couple of years.
Remote is hard. It’s challenging for three big reasons. First, if you are a Head of Engineering at a company like Coinbase, how do you build a global pipeline to find great people? If you want to hire tens of thousands, how will you build a pipeline of thousands of Golang engineers from Brazil or C-sharp engineers from Croatia? That’s hard.
Second, evaluating a global engineering talent pool can be tricky. For example, if you are looking at an engineer from Italy, say – you may not see Stanford or Berkeley in her educational background. You might not see Google, Facebook, or Stripe in her work experience. She could be a great engineer, but there’s just no signal from just the resume. So you have to interview that person. And how are you going to interview these thousands of people from all over the world without sucking up all of your engineering teams’ interview time — that’d be super hard.
And third, it can be tough to manage engineers effectively after you’ve found them. Because if your communication is a challenge because time zones are broad, often the right kind of daily communication and performance management does not happen. Often managers don’t have enough visibility into the work people are doing. Security can also be a concern.
So these are the three big problems.
Number one, it’s challenging to build a large enough global pipeline to find genuinely great people.
Number two, it’s hard to evaluate all these engineers at scale.
And number three, once you’ve found that perfect engineer, how do you manage them? How do you take care of communication, security, and other issues?
And the traditional solutions, Dan, weren’t built for this. If you look at a recruiting firm or a staffing firm, most of them don’t do any vetting of engineers. They don’t have a global reach. If you look at marketplaces, they’re usually hit or miss in terms of quality. And these IT services companies also don’t have Silicon Valley-caliber talent.
So we asked ourselves a simple question: Can we solve these problems with software?
What if we had software that could source engineers planet-wide? What if we had software that could evaluate engineers for a Silicon Valley bar? What if we had software that could automatically match the right engineers to the right jobs with machine learning? And what if we had software that could manage the collaboration after the match – this is why we built Turing.
Turing’s creating a new category called the ‘talent cloud.’ It’s a distributed team of developers in the cloud. These developers are sourced, vetted, matched, and managed by software. So an Engineering Manager, or the Head of Engineering, or an early-stage founder can push a button to spin up your engineering dream team in the cloud as easily as you spin up servers on Amazon. So that’s what we do.
This is great. Sounds excellent. So you Co-Founded Turing back in 2018, is that right? I’d love to hear what the journey to now has looked like for you and the team.
Yeah. It’s been a journey. It’s one of those things where every year feels very different from the year before. In the CEO role, as a company grows, you have to scale with the challenges you see at that next step.
Like in the last year, our headcount grew by almost 8X. So, the company’s been growing tremendously. A big inflection point for us was February or March 2020, when the pandemic hit. It accelerated a lot of this shift to remote work and this move to distributed teams. GitLab, Automattic, and a few companies tasted the benefits of remote, distributed teams.
And it’s never been a better time to be an engineer. Previously, your opportunity radius was maybe 20 miles from where you lived, regardless of how capable you were, how motivated you were, or how smart you were. Now that’s no longer the case. So at Turing, we want to kill the geo lottery.
So we want to create a future where the place you live doesn’t impact the kind of opportunities you can access. So these last three and a half, four years have just been this period of rapid scaling as we grow our developer base.
We now have about two million developers signed up on Turing and hundreds of companies building teams from Turing, including Coinbase and Johnson & Johnson. We also have fortune 500 companies like Disney and others. It’s been fun to experience blitzscaling in its purest form for all the good, the bad, and the ugly.
You raised $87m Series D in October last year – hit unicorn status. How are you investing the new capital? What do the next 12–18 months look like for Turing?
We raised a unicorn round of about 87 million last year. Since then, we have focused on scaling up our sales and marketing and accelerating our developer growth. There are many developers worldwide, and we want Turing to be where the best developers work. And really, investing in our product R&D. So, we build many products to automate the sourcing, vetting, and management of developers.
And it’s a lot of data science and machine learning coupled with software engineering to get the efficiencies of scale. So vetting is zero-touch in that we can have various job types, tech stacks, and seniority levels. So we want to build this machine that can evaluate engineers at scale in an objective, data-driven way without all the biases that a typical interview process would have.
Traditional interviewing is not very scientific. It’s kind of broken and has all sorts of room for bias. So we want to level the playing field for global talent. So a big focus for us after our series D is automation. So automation and sales and marketing, I would say, are the big levers. So we’ve raised about $140m so far. Most of the money is still in the bank as we continue to grow. We will also look at any attractive M&A opportunities in Europe and LATAM. So we’re always interested in great teams of people and technology that can give us an edge.
Forbes named you one of America’s best startup employers last year. Massive congrats on that. What is unique about Turing’s employee experience that you think secured this accolade?
So there are a few things that are unique to our culture. Our main pillars are speed, continuous improvement, and a long-term focus on customer success. And when I say speed, I think one of the biggest weapons a startup has is the ability to execute fast without bureaucracy and much red tape.
So we spend a lot of time thinking about how to go faster. And this means being very focused on what we do. It means only focusing on the significant needle-moving initiatives that can impact our metrics substantially. It means saying no to many things and saying no to a bunch of product initiatives that could be nice things to do but may not move the company forward in a meaningful way. And I think people like that culture. So when people come to us from some of these larger companies, they first notice the speed. This company moves fast.
One of the other attributes that fit in with speed is we are very comfortable failing. So we would rather take a big, bold bet in an area where we see an opportunity to do something 10x better or 5x better. And we’ll be happy if 80 percent of the time, rather than conducting tons of iterative, incremental improvements. And I think people like that. So that was speed.
And a second important part of our culture is a culture of continuous improvement. So when we hire people, we look for people who care deeply about making themselves better, making their teams better, and making the company better. It’s this mind mindset of getting better every day. I have an app on my phone where I track – did I work on an aspect of my self-improvement today? I like to maintain a streak of continuous improvements because these things add up.
So what will you focus on from a talent and people perspective in the year ahead?
So we’re focusing on hiring exceptional leaders in the company and ensuring that our team is coordinated, aligned, and moving in the right direction. We are now about 700 people. It’s much harder to keep an organization of 700 people focused on the most needle-moving things than when we were 70. So it’s going to be recruiting extraordinary leaders into the company.
So I would say recruiting and ensuring that the entire team focuses on the right things. Everybody has a clear sense of company priorities, their team’s priorities, and how they contribute to moving the company’s key metrics forward.
One thing that sometimes gets missed is something that’s in between recruiting and ensuring the organization is aligned and moving in the right direction, which is onboarding — making sure that we are making the people we hire successful. So we have the proper checkpoints with them — and that’s a whole different topic with its own challenges — you need an organization to have the right balance of leaders and individual contributors.
I prefer the term leaders rather than managers at Turing. Like we want leaders, not managers. We want people who raise the level of performance of their team, not somebody rubber-stamping the work of an incredible team. So, ensuring we have the correct ratio of leaders to individual contributors will be a focus. So, recruiting, excellent onboarding, and having a great culture where the entire organization is moving in one direction to hit our company goals.
You’ve also been named one of Fast Company’s ten most innovative companies in 2021. I’m not surprised, given the demand for tech talent and the global transition to remote work over the past couple of years.
Why do you think unlocking opportunities for global tech talent is so important right now?
Today, we live in a world where every company must become a software or a technology company to survive and thrive. Right. And the fundamental scaling constraint to a technology company is having great engineers and the ability to unlock the world’s untapped human potential. There are great people worldwide who could be the perfect engineer in your team to contribute to helping you go where you need to go. I think that’s the message that resonates powerfully with every tech company.
When you think about it, do you want to hire the best people in the world or people who happen to live near your office? It feels stark in terms of what the best path is. So we’ve benefited from those tailwinds. Traditionally, engineers that wanted to work in the heart of the technology industry had to relocate to a few centers in Western Europe, the west coast of the United States, certain parts of India, parts of China, or parts of Israel.
These are influential tech hubs, but you have to uproot your life and move to those places to work. And today, the jobs come to you. And I think we look back on this era as being transformative, much like the internet was in the nineties regarding how it connected the world and made civilization progress faster.
So we are fortunate to be at the center of that shift which is a big reason why they included us in many of these lists. We were named alongside Slack, Zoom, and GitLab. These companies are powering the boundaryless future, where you can work from anywhere. And that’s the movement; it’s the work-from-anywhere movement. And I think it will be a hugely positive movement for the world.
Even outside of the tech industry, if you look at it from an environmental standpoint, how much pollution are we avoiding by not requiring people to commute one to two hours daily? How much productivity do we lose when everyone commutes daily for one to two hours? And then they’re a little bit tired when they reach work. You’ve lost so many hours of your day – like 10 – 20% – it’s a significant amount of life that you’ve now got back to do whatever you want.
You held your event Boundaryless recently, didn’t you? What were some of the key takeaways from this?
We used that event to announce big product launches for us. For example, one significant product we launched in our Boundaryless event was a completely automated self-serve system that made working with engineers, like picking the right engineer you want to work with, as easy as going on Amazon.com.
Let’s say you are starting a company and want a backend Python engineer. So we have this system now where you can input what type of developer you’re seeking. What are the critical tech-stack strengths you need in a developer? Maybe it’s Python, perhaps you also want to add Node, and then you’ll see a ranked list of prevetted engineers from Turing, and you can push a button to choose which engineer you’d like to interview and get started.
And it’s just very, very efficient. We’ve taken a process that would typically take months and reduced it to a matter of days and, in some cases, the same day. And that took a lot of work behind the scenes to automatically evaluate engineers that scale used machine learning to recommend the right developers for the right jobs when choosing from a pool of 1.2 million. And we previewed that, and today, more than half of our startup customers use that product. And a lot of engineering managers value efficiency. Like they don’t like talking to a salesperson, getting on zoom. So they like this search engine just to find the developers they want, push a button, and get going.
What tech candidate assessment or evaluation do you feel is most effective for fast-growing companies?
So for fast-growing companies, when we evaluate software engineers, we assess them along three primary dimensions.
We evaluate their technical skills, we evaluate their soft skills, and we vet their seniority level. When we assess an engineer for their technical skills, we build a deep developer profile for each type of engineer. This profile is a detailed, comprehensive, continuously updating vector representation of that developer’s strengths and areas for improvement.
So with a machine learning engineer, we would evaluate them for how good their machine learning theory foundation is, whether it’s probability statistics, linear algebra, or things like that. We evaluate them for how hands-on they are and how good they are at building a text classifier and working with the latest frameworks. We would also examine their software engineering fundamentals.
How good are they at writing production-level code? We would evaluate them on their ability to build machine learning models versus maintaining the models in production. So we have all of these attributes that we are vetting the engineers for. Where relevant, we also evaluate them on things like their systems design capability or ability to architect systems and stuff like that.
And in the second bucket, soft skills, particularly for a startup, it’s crucial to have very proactive engineers with an ownership mentality who don’t need a lot of direction.
Typically in a startup, the engineer might report to somebody fairly senior, maybe one of the founders or a CTO or VP of engineering. So they need to be the kind of person who doesn’t need a ton of hand-holding where they can understand the vision for a feature or a product that you’re trying to build. They need the ability to take that to completion without requiring a ton of iteration and back and forth with the person they’re working with. They must be able to work with minimal supervision and be committed to working hard. I mean, startups are hard work, right?
Like it’s not for everyone at all stages of their life. So you kind of want somebody who’s committed to the company’s mission and can work and put in those long hours. Look for somebody good at direct communication and escalating when things are not going right. In a startup, speed is paramount. So you don’t want somebody who sort of says yes to you.
You want somebody who negotiates more directly with you. So the soft skills front, particularly for somebody working at a startup, I think some of these requirements are important in a startup, as you might also need to wear a couple of different hats.
Sometimes the engineer might need to wear a more product-centric hat too. You might have to make some product-centric decisions. You might have to work with a designer. You might have to talk to customers. So you kind of need all of that too.
And on the third dimension, we pay a lot of attention to the level of seniority the client seeks. We have engineers who can work at the level of a task, at a feature level or the level of an entire product. And we typically have a conversation with our startup or enterprise customers to understand what seniority level they need.
So it’s technical skills, soft skills, and calibrating on seniority levels so that we can help companies find the right talent they need. And often, Dan, it’s a conversation. Sometimes when customers come to us, they have a vague sense of what they need. And in a conversation with us and through iterating with our product, we help them sharpen their job rec for the task that they need to be done.
Sometimes, what you need for a project might not be a machine learning engineer. Instead, it’s more of a data scientist or maybe an engineer who understands data sciences – it’s an iterative process to figure out what our customers need.
And aside from technical challenges, how do you assess for qualities like culture fit / add or hiring for “potential” even?
Yeah. How do we assess for culture fit and hire for potential? So culture fit is something tricky. Let me answer that first from a perspective, and then we can talk from a customer perspective. I think it starts with the Founder and CEO writing down the culture. So I spent some time writing down all the best practices from the culture we want, treasure, and value in the last month. We call it the Turing way, and we’ve written it down in this Google doc in conversation with our exec team – what traits have made us successful so far that we want to preserve?
So it starts by writing it down because different companies have different cultures, and there’s no one size fits all, but you have to write it down to put a stake in the ground for what you stand for. And you’ve written it down.
Then you need to have a way to hire and fire and promote based on those values. So one of the things we are doing now is we’ve written it down, and we write specific examples of what each sort of cultural value means, and some of it can be kind of polarizing. Like in our culture, we write that we work crazy hard.
We think Turing will be one of our generation’s most influential companies to unleash the world’s untapped human potential. And it’s going to take a ton of work, and we want you to know what you’re getting into, right? So this is not a company where things will go slow, and it’ll be a lot of work, but we can promise you it’ll be rewarding and fun.
So the first step is to write it down. And you also want to write it down collaboratively, like taking input from the outstanding leaders you have in the company and your exec team. Then you also want to be mindful of not being too ossified in the culture itself. Someone told me that they hire not for culture fit but for culture addition. So you want to hire people who will contribute positively to the culture and improve the company’s culture and who can add their own to the Turing way.
So we kind of watch for that in our interview process. We have people who interview for culture. You want to have like a very standard way in which you assess culture. Ask the same questions to determine your ability to contribute to Turing’s culture. So you calibrate it across a broad group of people and share this culture document you create with prospective candidates and managers. Not have this be something sitting in Slack or a Google Drive, but actively use it. I think the more often you can point to that, the more it’s being used.
So step one is to write it down. Step two, be comfortable with evolving and editing it. And step three, have a hiring, firing, and promoting system based on what you’ve written down. So yeah. A culture document without enforcement is kind of toothless, right? So it’s important.
You have a database of 2m developers across 10,000 cities. That’s an incredible trove of talent. Talent that typically is in demand and tough to hire.
How have you attracted and engaged that talent to the extent that they opt-in to Turing?
Great question. So firstly, we live in a remote-first world, and every company’s in the race to hire the world’s best remote talent, but it can be hard to stand out in a planetary pool. If you are an excellent engineer from a small town near Sao Paulo, Brazil, nobody looking at your resume might recognize the schools you went to or your prior work experience. And that’s a shame like this could be a perfectly fantastic engineer, but there are just not too many signals that exist.
And if you’re an engineer historically before Turing, you had, I would say, three options that you could have, you could have done. One is you could have applied directly to the best companies that are hiring. And most of the time, when folks do that, they don’t hear back. So you get lost in the shuffle. Like you cannot get career growth, you kind of get mentorship. These are long-term engagements, and you’re working on exciting products. So that was good about them. The hard part was you never heard back. It’s hard to get noticed.
On the flip side, there used to be these marketplace companies, which were easy to get on, and the jobs were easy to get. You can post a job on some of these marketplaces. You might get a gig here or there, but these are gigs, not real jobs, like not jobs that contribute to your career growth. You don’t get good mentorship. Often you’re not working on the most important part of the product. You might be working on something on the site that people don’t care about much.
The third used to be. You could go work for like an IT services giant. The good thing is these are easy to get. But you’re not directly working with the companies. You’re working for the middle person. So we thought we could create a new category of work where we give people the benefits of each of these without none of the cons.
What if you had access? What if you could work for Coinbase or Rivian? You could work for Johnson & Johnson directly on their core products and have long-term engagements, career growth, community, etc. It’s a model that combines all of the benefits with none of the cons. Have you had the flexibility to take time off between engagements? Why we build – to satisfy that goal.
And people value the work we do on our community side to help our engineers uplevel their career growth. We have programs where we help them learn how to interview better, work on their soft skills, and work on their leadership skills – recommending what skills are in demand that they could learn to grow their value in the industry to get promoted faster.
So we want to give people like this guidance to be the sort of jet pack on their back to help them reach the heights they’re truly capable of achieving.
What have some of your biggest learnings been regarding building teams?
I would distill this into three significant learnings. First, building teams is being intentional about the job description. You are starting to hire for a role – it may sound obvious, but often the most significant times when we’ve had challenges is when that initial job spec wasn’t super clear in terms of what we were looking for in this role in this person. So spending time being clear on who we are looking for to do what role and how we will measure success, making sure that’s defined very well upfront. So that’s number one.
The second learning is that I would go back to our earlier chat on hiring for culture. And it only starts when you write it down. It’s not OK to just look for the person to do the job. Are they going to be a good culture fit for Turing? Do they get on well with other people on our team? We also look to do at least two back-channel reference checks for every exec-level role we hire. I think that that is important.
And the third learning for me is to –particularly for leadership and executive level roles – stay very close to the person for the first two to three months. I try not to give them a ton of responsibility too quickly. And it’s one of those things that requires a little bit of a lack of a better word, a top-down push where the person, particularly in leadership roles, I feel like the person may feel like they are ready at a particular stage sooner than you may know they are.
And there would probably be a phase where the person probably feels like they have sufficient context, but you have more context about what they know to kind of stay close to the person for the first two to three months to ensure they’re successful, I think is key.
So number one is being clear on the job description and what you require them to do. Then, how are you going to measure success? The second is making sure that a good fit for the culture and stage of the company. And third is, staying close to the person in the first two to three months.
Regarding the culture piece, I think I will be hiring a hundred inferior versions of this person. It’s going to be an army of this person. So the leader often becomes the ceiling for the function. So are you hiring people with a high enough ceiling so that they can attract amazing people? And the leaders also model what good performance is to their organization.
So if you have like a hundred people who will be lesser clones of this person in terms of their ability to contribute, would you be happy with that? I think that’s a sound check.
You build teams by building leaders, by hiring leaders. So you want to make sure that the leaders, the template of the leader, is precisely the one you want a lot of copies made in your company.
I like that. And if there’s one thing you could wave a magic wand at and fix when it comes to building and leading tech teams – what would that be?
My advice to any founder is that hiring speed matters with tech teams. And you’re going to hire the right engineering team fast by casting a planet-wide net. So I find companies sometimes being overly restrictive regarding where they hire, which always hurts the company. Your goal is to build a product that makes your customers happy and moves the metrics for the business.
And you want to do that as fast as possible so that you can grow as quickly as possible. And the biggest stumbling block can be the speed of hiring. If you’re hiring one engineer a month when you should have been hiring five, that can profoundly impact your ability to compete with companies with better hiring velocity.
So my advice would be to be very thoughtful about which countries your team is in. A 4-hour time zone overlap is all that’s needed for an engineering team. If not, you are losing out on great people unnecessarily because a lot of engineering time is spent inside a code editor, GitHub, Slack or JIRA, or tools like that. So a 4-hour overlap for tech teams is acceptable.
If you look at crypto or the open source movement, it’s a testament that distributed teams work with primarily asynchronous collaboration. So my advice would be to look for a 4-hour overlap. Other than that, cast as planet-wide net as you can so that you find genuinely great people. Tech teams are no different from other teams in that what makes a great team is a great leader who sets the right culture.
As the founder, you can’t ensure that every IC on the team performs at the level they need to. You need great leaders. So I would recommend having an engineering manager, director of engineering, or head of engineering who’s hands-on.
At Turing, we have a culture where the leader is usually the best engineer in the team and is also good at managing the team. That way, they can unblock their team, which can help them identify and make the right architecture, systems, and design decisions so that the company moves critically in the right direction.
So I would say the two pieces of advice would be to cast a wide net, have a 4-hour overlap for distributed teams, and ensure you have a hands-on engineering manager. I think for a tech team today, I would split it into engineering product and data science; usually, design is in the product. You also want a tech team collaborating well with their peer organizations.
You want an engineering team that can be an excellent partner to product and a strong partner to data science. You want a data science team that can be an exceptional partner to product, a good partner to engineering, and data science is a somewhat new function that didn’t exist in this form maybe five years, even ten years ago, for sure. Perhaps even five years ago. So it’s essential to clarify the boundaries for who makes what types of decisions, like, what is data science responsible for? What is engineering responsible for? What is product responsible for?
And when you are building a team, you might want to think through that more carefully. This new tech category of data engineering is now different from data science at Turing. We have data engineering under data science. As a result, we need to be very thoughtful about who defines the data layer in a company, which defines the database, which defines the nature of the metrics being tracked, the events being logged, the schema for a dashboard that you build, and who determines where the database sits, where the servers are located.
So you kind, when you’re building a tech team, you want to be thoughtful about not just your team but your engineering team, your data science, data engineering team, and your product team, and how you split responsibilities between them.
A couple of light-hearted questions to bring our chat to a close today. First, is there anything you’re super passionate about? Something you find unapologetic amounts of joy in – this can, of course, be professional, personal, or both!
Thank you, Dan, for that question. One thing that gives me joy is working on my continuous improvement. I want to wake up a little better every day than I was the next day. So I have a long list of areas for that. I want to up-level myself in places I want to improve at that I actively work on. And I always feel happy when I work on something related to my improvement. And these things compound over time. there’s a lot of value in it.
On the personal side, spending time with my wife and my one-month-old daughter is a huge source of joy and fun. Besides that, I love being at the cutting edge of machine learning. So I love reading up and playing with some of the most recent machine learning frameworks, just building things for fun.
That’sat’s interestiIt’sIt’s been so great chatting with you today. I enjoyed our conversation. I appreciate the time. So thank you very much for being on our podcast.
Likewise, Dan, I enjoyed our conversation as well, and for Founders, there’s no better time to build a startup from anywhere in the world. You can fundraise from anywhere. You can hire a team from anywhere. It doesn’t matter where you are based anymore. So I wish you the best of luck in building your companies.
And if you need to hire engineers, do check out Turing!
Featured Image Credit: Provided by the Author; Thank you!