API

'Success of company turns people into stars, not vice versa': Thuan Pham, Uber CTO

Harshith Mallya
11th Mar 2016
  • Share Icon
  • Facebook Icon
  • Twitter Icon
  • LinkedIn Icon
  • Reddit Icon
  • WhatsApp Icon
Share on

Just a week after soft launching ‘UberMoto’, a bike taxi service in Bengaluru, Uber India on Thursday announced that it had opened its first Asian engineering centre in Bengaluru, to scale up its India hub and use it to solving customers' problems worldwide.

On the backdrop of this visit, Thuan Q. Pham, the CTO of Uber, spoke at length and mesmerised the crowd at NASSCOM’s Warehouse in Diamond District, Bengaluru with Ravi Gururaj, Chair of the NASSCOM Product Council, moderating the session. Here are some of the major highlights from the session.

YourStory-Uber-CTO

Growing up

Thuan’s family was originally based in Vietnam, but had to leave the country after the Vietnam war. His mother took him and his brother out of the country on a boat and after spending some time at a refugee camp in Malaysia, finally landed up in USA in 1980.

Thuan faced some difficulties adjusting initially as he was in a new country, didn’t know English and had to start from scratch. He said,

Things got better and I was fortunate to make it to Massachusetts Institute of Technology (MIT) and earn undergraduate and graduate degrees in Computer Science.

Soon after that he landed an internship with HP research labs in the heart of Silicon Valley and has been at the Valley since then because of his love for technology. Like most parents, Thuan’s mother expected him to stay at his cushy research job at HP for many decades and she was quite alarmed when he decided to leave his secure job for smaller startups. Thuan said, “I like to constantly learn, so I didn't want to allow myself to get comfortable at one job. I wanted to build things people actually used, rather than doing research and other intellectual pursuits.”

So over the course of his career, Thuan worked at Silicon Graphics, DoubleClick, and VMware. Recalling his days at Silicon graphics, he said that although the company had an amazing team and though they were working on ground-breaking technology, like video streaming and online gaming, they were not able to take off as the end costs were too high and they were too early in the market. He said,

We learnt a valuable lesson that intellectual prowess is not everything.

Joining Uber

After working at VMware for a long time, Thuan was tired of working at a large organisation and wanted to work on something new. So he thought of taking a year off to relax and spend more time with his family. It was at this time that Bill Gurley, General Partner at Benchmark, requested Thuan to consider interviewing for a job at an upcoming Silicon Valley startup – Uber.

After meeting Travis Kalanick, Founder of Uber, for almost two hours, Thuan was impressed by what he saw. He ended up talking and interviewing with Travis for almost 30 hours, over the course of many days, before finally agreeing to join Uber. Thuan said that the two of them had some really challenging conversations across a variety of topics.

I had never done an interview like that before. It was great fun.

Related read: 9 ‘geek’ lessons to entrepreneurs from Travis Kalanick, CEO, Uber


At the time, Uber was still in its early days and active in about 30 cities with 200 employees. But Thuan saw a cultural fit and liked the challenge. So he decided to join Uber as it met the three most important criteria that he judged company’s on.

  1. Work has to be impactful: Should be a big challenge and opportunity to make an impact.
  2. Like the team: One spends more time with the team than family during weekdays, so it is important to see a cultural fit.
  3. Founding team: “You only get to choose your boss, one time. So choose wisely.”

Looking back at his entire career so far, Thuan said that he hadn’t really planned his career trajectory but had just focussed on the work at hand and looked at bigger challenges. He feels that one should not feel safe and secure because of past successes, He said,

We are only as good as what we can do tomorrow.

The secret sauce at Uber

Thuan admitted that though Uber is not perfect by any means. They have been able to accomplish a lot because of the culture of the company built by the initial core team and their ability to execute and scale rapidly. He said,

We generally make the right decisions at the right time and that is our secret sauce.

Talking about what he looks for while hiring people, Thuan considers passion, willingness, and the ability to learn quickly as top qualities. He said, “It is very easy to tell when people are passionate about what they are talking about. You don’t need to be an expert, but should be able to talk about certain topics with certain depth. It is easy to recognise posers, who only pretend to know something.”

Talking about the signs of a great team, Thuan said that most of the best people are self-driven and work because of passion for the problem, rather than the paycheck. He said, “The best people know they can draw amazing salaries anywhere they want to. So while working at a young startup, they look for companies that are worthy of their time, which pose challenging problems.”

Thuan admitted that like every big company, they have also lost a lot of great talent over the years because they were not able to create the ideal culture for all of them. He said,

Set an environment or culture where everyone cares.

He also believes that everyone should be accountable to themselves and not pass the buck. He said, “The head of sales shouldn’t complain that he or she can’t hit their quarterly goals because a certain feature is missing. Instead one should focus on improving and learn how to sell ice to eskimos”

Going from 40 to 1,200 engineers

While Uber had got the necessary product-market fit early on, they faced big challenges when it came to scaling globally at a rapid pace. Thuan said, “To meet demand, we went from 40 to 1,200 engineers over the course of three years.”

Thuan recalled that Travis made a strategic decision to go international very quickly. Looking back now, he considers it a very smart decision though it added more complexity. He said,

We had to live with the reality of being in multiple markets simultaneously. Engineers needed to solve tough global problems they were not exposed to. We had to grow up that way and it helped a lot in the long run.

Talking just about India, Thuan added how Uber realised they need to adapt to unpredictable environments a country that had network drops and slower 2G internet services back in 2013, when they were exposed to a more stable environment in the US.

As they scaled, the company also needed to hire more experienced engineers to solve world class problems. Thuan found that Uber went from hiring senior engineers with five years of work experience in 2013 to engineers with nine years of experience in 2015.

At the end of the day, Thuan feels that team size doesn’t matter. He claimed that though Uber has a smaller team compared to its peers, it has been able to achieve more because of its culture and quality of people. He said,

The company needs to be highly aligned – but loosely coupled at the same time. You need to grow big but still have the agility of a small company.

On how they make important decisions, Thuan said that though many people come with different perspective, there are debates but at the end of the meeting the best idea wins and the whole team backs it irrespective of their personal opinions.

Technology and debt

At Uber, each engineering team has their own roadmap that don’t generally overlap. They run on a variety of programming languages and give engineers freedom to an extent on their choice of language. Uber also tracks a lot of data points from customers and relies on it to predict where cabs will be needed based on past patterns. Their dream goal is to bring down the estimated time of arrival (ETA) of cabs to zero minutes in the ideal scenario.

Talking about ‘technology debt’, Thuan said that it hadn’t changed much over the past few decades. He said that the strategy had to be to move fast and cut corners in early stages of developing a product. He said,

The debt you pay is inefficiency. It is ok to accumulate some tech debt but you have to constantly budget time to fix it.

Culture and career graph

YourStory-Uber-CTO-2

Thuan believes that every company should clearly state their culture and values. “Travis was instrumental in building the culture from our early days. It makes things easier for new employees as they are able to look around and take it in quickly.”

To make onboarding easier, Uber has an internal team that provides specific training such as ‘Engucation’, a combination of engineering and education, and ‘Uberversity’, a four-day crash course training course.

Thuan also personally looks after onboarding new employees. Every Monday morning, he gives new employees a talk for 90 minutes to help them get on the same page and ensure they know their roles. Uber has a ‘dual career’ graph(for engineering and management) that they follow to help employees visualise their long-term career growth at each step.

Growth hacking

The old school thinking style dictates that startups should hire the most experienced people but they can have egos. Also convincing someone to join a ‘fly-by night company’ with an unproven business model is not easy. So startups need to go with who they can find easily in the early days. Thuan said,

All early engineers are generally unproven and not household names. Success of company turns people into stars, not vice versa.

On the subject of software development methodology, Thuan believes in keeping it simple and agile, he said, “We just wing it. No management author really understands the situation your company is in. So don’t focus too much on following them.”

On dealing with customer feedback, Thuan said that though it is tempting to try to solve all the customer inputs, the focus in early days should be on the main problem. The thought process should be,

What is problem if I don’t solve today, will kill the company three months on.

With the success Uber has achieved, he now has more breathing space. Thuan’s thinking has now changed to,

What is problem if I don’t solve today, will kill the company one year later.

Engineering centre in Bengaluru

On being asked, why Uber had opened an engineering centre in Bengaluru, when it already has many in other cities. Thuan recalled an incident from his early days at Uber. He had gone on a holiday with his wife and they took an Uber to reach their destination in the city. At the end of the ride, his wife marveled that the system was already perfect and working and questioned him what he was currently doing at the company.

He tried explaining but couldn’t get the point across in engineering terminology. So he explained the importance of continuous scaling with a house meant for five people, needing to constantly accommodate more and more people. He said, “Imagine, the toilets get clogged, plumbing suddenly stops working, rooms are not enough. That’s what is happening.”

Uber’s long-term vision is to serve customers all over the globe and it sees India and China to be important markets. “Only way any of us get better, if we go above our expectations and boundaries from time to time. We need to seek discomfort, risk and failure as it helps, stay on course.”

The future

Ravi asked Thuan about the future course that Uber is taking and what they consider Uber to be at its core. Thuan said “Our long-term vision is to enter all kinds of demand markets. We are moving consumer goods through UberEat and UberFresh and so aim to solve anything in this realm.”

On the topic of driverless cars, being a realist, Thuan feels that it may become a reality only 10 years or so down the line because of the inherent risks, regulatory concerns, and the sheer engineering challenges involved. So for a short-term, Uber is mainly focussing on the human driver-car problem and has driverless cars on the backburner.

Thuan believes that in the future it may be possible for tech companies to take people from point A to point B across different modes of transport on-demand. Citing an example of travelling from Bengaluru to San Francisco, he said that technology may be able to provide on demand service, the moment we need it at every leg of the journey, across different modes of transport.

  • Share Icon
  • Facebook Icon
  • Twitter Icon
  • LinkedIn Icon
  • Reddit Icon
  • WhatsApp Icon
Share on
Report an issue
Authors

Related Tags