How Shopify builds a high-intensity culture | Farhan Thawar (VP and Head of Eng)

Farhan’s background (0s)

• Choosing the harder path can be beneficial, even if it doesn't work out, as it allows individuals to work with smart people, learn something valuable, and gain experience (3s).
• Instead of just working more hours, creating intensity in an organization can be achieved by doing more per minute and simplifying processes (27s).
• Shopify has a "delete code club" that aims to simplify code and has found over a million lines of code to delete (37s).
• Asking questions, even if they seem stupid, is important for understanding content and can be a superpower (42s).
• Farhan Thawar has worked for different billionaires every decade of his life, who share a common trait of being similar in one thing (1m7s).
• Shopify is a fully remote company with over 10,000 employees that operates with urgency, velocity, and a first-principles way of thinking (1m29s).
• Farhan Thawar maintains intensity and urgency within the engineering team through meeting cadences, pair programming, and canceling meetings (1m47s).
• Indexing towards choosing the harder option can make life easier in the long run (2m3s).
• Farhan Thawar shares hiring advice and stories, including the importance of looking for people who are willing to learn and take on challenges (2m7s).
• Shopify's engineering intern program plans to hire over a thousand engineers in 2025 (2m15s).
• The concept of building a high-intensity culture is discussed, drawing parallels with Steve Jobs' perspective that everything connects when looking back in retrospect (3m32s).
• Measuring engineering organizations can be challenging, and simple metrics like the number of PRs or commits do not provide a complete picture (3m48s).
• DX is an engineering intelligence solution that combines quantitative data from developer tools with qualitative feedback from developers to give a complete view of engineering productivity and the factors affecting it (4m2s).
• DX is designed by leading researchers, including those behind the Dora and SPACE frameworks, and is used by iconic companies like Etsy, Dropbox, Twilio, Vercel, and Webflow (4m0s).
• Persona is an adaptable identity platform that helps businesses fight fraud, meet compliance requirements, and build trust by securely verifying individuals and businesses across the world (4m47s).
• Persona offers flexible building blocks that allow companies to build tailored collection and verification flows, maximizing conversion while minimizing risk, and its orchestration tools automate the identity process to fight rapidly shifting fraud and meet new waves of regulation (4m57s).
• Persona is used by leading companies like LinkedIn, Etsy, and Twilio, and offers plans for both startups and enterprise businesses (4m52s).

Choosing the hard path (5m38s)

• Choosing the harder path in life can make it easier down the road, as it allows for learning and growth, even if the chosen path doesn't work out (5m40s).
• If the easy path is chosen and it doesn't work out, it can result in not learning anything and not working with smart people, ultimately leading to a loss of choice (6m10s).
• On the other hand, choosing the hard path and it not working out still results in learning and growth, as well as working with smart people (6m30s).
• An example of this is job seekers who send out numerous resumes a day, but don't learn anything from the process, versus building something on top of a platform like Shopify, which can lead to learning and potentially getting a job (6m41s).
• Taking hard courses can also lead to meeting smarter people and learning more, even if it results in worse marks (7m27s).
• The people involved in the hard path are crucial, as they can provide a valuable learning journey and lead to intense working relationships that can continue throughout life (8m18s).
• It's essential to consider whether the hard path is also a smart and worthwhile one, rather than just being difficult for its own sake (8m13s).
• The hard path should be chosen because it provides an opportunity to learn from smart people, be in a constant state of growth, and develop skills that can be valuable in the long run (9m2s).
• The goal is to choose a hard path that allows for learning and growth, rather than just doing something difficult in a dumb way (9m22s).

Getting comfortable with looking dumb (9m37s)

• Embracing the idea of looking stupid in public and trying new things is considered a superpower, and it's essential to make it a team superpower as well (9m38s).
• Trying new things can sometimes lead to disaster, but it's crucial to have guardrails in place to minimize risks and learn from failures (10m26s).
• Being okay with failing and looking dumb is a hard skill to develop, but it's essential for growth and learning (10m58s).
• Growing up working in retail and dealing with rejection from customers helped build confidence and develop a resistance to negative reactions (11m16s).
• Telemarketing, cold calling, and other activities that involve rejection can also help build up resistance and confidence (11m55s).
• Not caring about looking dumb and being willing to ask questions, even if they seem stupid, is key to learning and growth (12m8s).
• Having a growth mindset and being able to bounce off negative reactions is empowering and essential for success (12m39s).
• Even successful people face rejection and criticism, but it's how they respond to it that matters (12m28s).
• The goal of asking questions, even if they seem stupid, is to understand the content and learn, not to annoy others (13m5s).

Lessons from working with visionaries (13m20s)

• Farhan Thawar has worked with multiple successful individuals, including Joe Lemond, Chamath, and Toby, each in a different decade of his life, and notes that despite their differences, they share a common trait of having an irrational view of what the world should look like in the long term (13m29s).
• These individuals are long-term thinkers who can envision the future 10-25 years ahead, and Farhan recognizes that he is not good at seeing this vision, but is good at moving towards it 1% a week (14m4s).
• Farhan believes that the key to success is finding people with this irrational vision and merging forces with them, as he is too reasonable to come up with such visions on his own (14m42s).
• He has actively sought out these unreasonable people, including when he started his own company in 2015, and met with entrepreneurs in Toronto to find someone to work with (14m58s).
• Farhan emphasizes the importance of recognizing one's strengths and weaknesses and doubling down on strengths, rather than trying to build a superpower that is not natural (15m43s).
• He also mentions that having a framework for what he cares about has changed the way he picks jobs and helps him avoid getting confused by things that are not on his framework (15m53s).
• A personal framework helps in making career decisions and avoiding distractions, as it outlines the things one truly cares about, making it easier to determine if a job aligns with those values (16m26s).
• Having a written framework saves time by preventing distractions and forces one to reevaluate their current situation to ensure it aligns with their values (16m51s).
• The framework consists of questions to ask about a potential workplace, helping to determine if it's the right fit, and can be used as inspiration for others to create their own (17m36s).
• Resigning from a job that doesn't meet the framework's criteria is a viable option, as seen in the example of resigning from a previous company due to a change in role and responsibilities that no longer aligned with personal values (17m45s).
• The previous company, Extreme Labs, was a mobile app development company that worked with major brands, but after being acquired by Pivotal, the role changed, leading to a misalignment with personal values (17m50s).
• The experience of resigning from Pivotal led to new opportunities, including becoming an investor in new companies that spun out from there, and was ultimately a positive outcome (19m7s).

Creating intensity in organizations (19m19s)

• Creating intensity in an organization is crucial as it allows individuals to expend more energy per hour, rather than just spending more hours on work, which can lead to extraordinary outcomes (19m32s).
• One hour is the same for everyone, and if individuals can get more done in that hour, they can achieve more without having to work longer hours (19m55s).
• This approach enables people to maintain a better work-life balance, as they can complete their tasks efficiently and still have time for personal activities (20m21s).
• An example of an intense activity is pair programming, where two people work together on one machine, focusing on writing code without distractions, which can lead to high productivity but also exhaustion (20m46s).
• People who are extremely good at their job can produce high-quality output quickly, which means they can achieve more in a shorter amount of time compared to those who are just good (21m33s).
• This concept is related to Parkinson's law, which states that work expands to fill the time available, but in this case, it's about achieving more in less time without compromising quality (21m56s).

The power of pair programming (22m6s)

• Pair programming is considered the most underutilized management tool in engineering, where two people work together on one computer, using tools like Tuplo for remote collaboration, and it is not about writing lines of code, but rather finding an elegant solution to a problem (22m35s).
• The throughput limiter in pair programming is not the hands-on keyboard, but rather thinking through the problem and building the right solution, and it is not about writing more code, but rather finding the best solution (23m5s).
• Toby, a developer at Shopify, used to pair program with his CTO Cody, setting a timer for one hour, and if they didn't finish the problem, they would delete all the code and start over, keeping only the tests, to ensure they were building the right thing (23m24s).
• This approach may seem extreme, but it ensures that the right elegant solution is found, and it is often better to rewrite code immediately after writing it, as it is still fresh in the mind, and it allows for pathfinding and finding a better solution (24m12s).
• Code is a liability and lives a long time, so finding the right solution is crucial, and pair programming allows for high rates of learning, as individuals can learn from each other's keystrokes, problem-solving approaches, and thinking (24m57s).
• While pair programming may result in writing less code, it allows for faster delivery of value to customers, and it is a powerful tool for finding the best solutions to problems (25m14s).
• Studies have shown that happiness, knowledge transfer, and intensity are higher when working in pairs, with a lower likelihood of silos, at a cost of around 20% of what would normally be done, similar to the underhanded free throw in basketball which statistically scores more baskets but is rarely used due to its appearance (25m19s).
• The underhanded free throw is an example of something that is effective but not widely used because it looks dumb, as seen in the case of Shaquille O'Neal, a Hall of Famer who was notoriously bad at free throws and refused to use the underhanded method despite its potential benefits (25m32s).
• Pair programming is a method used at Shopify, where two people work together on a problem, and it is used in a flexible way, with some teams doing it for 4-8 hours a week, and others using it as a tool to move quickly down a path, but also spending time on pathfinding and trying to figure out what to build (26m16s).
• At Shopify, pair programming is not used all the time, but rather as a tool in the arsenal to move quickly, and it is often used when working on important segments of the codebase or during incidents (27m34s).
• Pair programming is also used when working with someone to figure out a problem together, and it is a way to learn and gain knowledge, with the goal of eventually deleting everything and starting over with the new knowledge gained (27m29s).
• The use of AI, such as GitHub co-pilot, is changing the way people work, and it can be seen as a pair programmer, allowing people to feel like they are pairing without another human (28m5s).
• The use of AI in coding is becoming increasingly prevalent, with tools like Whisper allowing developers to communicate with the cursor and build new components using voice commands (28m18s).
• This technology enables developers to go back and forth with their pair programmer using only voice, eliminating the need to type (28m32s).
• However, it is suggested that the ideal coding experience would involve a combination of AI co-pilots and human collaboration, allowing developers to generate code, review, and refine it together (28m35s).
• The integration of AI co-pilots in coding has numerous benefits, including the ability to try new programming languages with ease, as the API and syntax become more accessible (28m59s).
• The presence of AI co-pilots also means that developers never have to code alone, making the coding process more collaborative and efficient (28m58s).
• The advancements in AI-assisted coding are seen as a positive development for engineering and software development as a whole, making it more accessible to everyone who wants to build software (29m10s).
• The future of coding is likely to involve a significant amount of pair programming, with AI co-pilots playing a crucial role in the development process (29m16s).

Shopify’s culture of intensity (29m18s)

• Shopify uses a tool called GSD (Get Stuff Done) which involves weekly updates to the whole company on what's happening, encouraging teams to show progress every week (29m35s).
• The company has shifted from having twice a year cadence to six week reviews, where teams come together to discuss their roadmap, resourcing, and progress with their immediate leadership and Toby (30m4s).
• The six week review process creates intensity as teams want to show what has been done and what they've learned since the last review, and it's a good cadence because it's short enough to remember context and long enough to accomplish significant work (30m40s).
• The company doesn't wait until the next six week review to implement changes, instead they start building and iterating the next day after receiving feedback (31m4s).
• Shopify applies Parkinson's law, which states that whatever time is allotted to something will be the time it takes, by moving quickly and getting things done in a short amount of time (31m16s).
• The company doesn't do change management, instead they "land" changes and then inform everyone, which allows them to get things done quickly and move on (31m50s).
• Shopify's approach to change is influenced by its volatile nature, and the company is willing to make changes quickly, even if it's not smooth, in order to move forward (32m7s).
• Shopify's culture emphasizes moving fast and being resilient to change, even if it's not smooth, as it's better to make changes now rather than waiting (32m16s).
• The company uses a "chaos monkey" approach to test its systems' resilience by intentionally causing disruptions, similar to how Toby increased resiliency in the company by unplugging machines in the data center (32m29s).
• Shopify prioritizes speed and delivering value to merchants over having a long change management roadmap, which would hinder progress (32m46s).
• Weekly updates are shared among team members, where each person discusses their current projects, and projects can include videos and written updates on changes made since the last update (32m54s).
• Shopify uses a process called "ok1" and "ok2" for project alignment, where ok1 is typically at the director level and ok2 is at the VP level, to ensure projects align with the overall direction and architecture (33m8s).
• Every six weeks, the team reviews progress with Toby, who encourages them to move forward quickly and not overthink things (33m35s).
• An example of this intensity is when the team was trying to solve an AI problem with large language models (LLMs) and needed a large output context window, but most LLMs had small windows (33m54s).
• By asking the LLM teams directly, Shopify was able to increase the output context window on major models within an hour, allowing the team to move forward quickly (34m17s).
• This approach increased the team's intensity and allowed them to get unblocked quickly, demonstrating the importance of asking questions and moving forward rapidly (34m51s).
• Regular check-ins and meetings are used to create intensity and velocity within Shopify, which may seem counterintuitive to the common idea of having fewer meetings, but it helps ensure that people are not working on unnecessary tasks and allows for refinement of priorities (35m1s).
• The goal of these check-ins is not to micromanage or catch people not doing their work, but rather to pair on problems and provide context to move forward, with the intention of working together to solve problems (35m37s).
• This approach is based on a combination of trust and verification, where the goal is to collaborate and provide advice, rather than simply monitoring progress (35m38s).
• The idea of micromanagement is not seen as a dirty word at Shopify, but rather as an opportunity to work closely with team members and provide guidance and support (36m23s).
• This approach is similar to the concept of pairing, where two people work together on a problem, and it allows for a deeper understanding of the work and the ability to provide context and advice (36m49s).
• This approach was learned early on in Shopify's tenure, where the CEO, Toby, would have one-on-ones with team members to work on problems together, rather than just assigning tasks (36m53s).
• The intention behind these check-ins is to work together to solve problems, rather than just assigning tasks or monitoring progress, and it requires a different mindset and approach to management (37m10s).

Meeting Armageddon: revolutionizing company meetings (37m18s)

• Meeting Armageddon is a process where all recurring meetings with more than two people are deleted once a year at a random time, followed by a two-week moratorium on adding new recurring meetings, to rethink meeting necessity and free up time for individual contributors (37m49s).
• The process, also referred to as "chaos monkey," forces teams to reevaluate their meeting cadence and consider alternative approaches, such as one-time meetings or different frequencies (38m31s).
• The deletion of recurring meetings and the migration of Slack to Workplace Facebook resulted in a significant decrease in the time individual contributors spent in meetings, from around 5-6 hours per week to 3 hours per week (39m3s).
• This reduction in meeting time led to productivity enhancements, allowing individual contributors to have more "flow time" to work on tasks (39m8s).
• Managers also saw a decrease in meeting time, from around 10 hours per week to 6-7 hours per week, which is still considered reasonable for alignment purposes (39m22s).
• The idea behind Meeting Armageddon is to combat the inertia behind recurring meetings, which can be difficult to delete due to their routine nature (38m12s).
• The process is inspired by Dropbox's approach to meetings, which was shared with Toby prior to its implementation at Shopify (37m41s).
• Meeting Armageddon is a unique approach to meeting management that can be learned from and applied to other companies (37m26s).

Reducing distractions (39m46s)

• Moving announcement information from Slack to Facebook Workplace (now being deprecated) significantly reduced distractions, as it shifted company updates to a feed that can be consumed differently than Slack's messaging system (40m1s).
• The goal was to discourage discussions on updates that didn't require them, and while Workplace allows commenting, it's not the same as using Slack for this purpose (40m47s).
• Slack is still used and valued, but it wasn't the right tool for company updates, which is why the information was moved to Workplace (40m55s).
• The ideal tool for company updates is likely a "river of information" that can be dipped into, rather than a chat system that sends alerts and encourages discussions (40m36s).
• The company is now looking for a new tool to replace Workplace, as it is being deprecated (40m12s).

Deleting 1M+ lines of code (41m10s)

• PA programming is considered the number one factor in creating intensity, as it requires individuals to be fully focused on their computers. (41m26s)
• The weekly cadence, which is part of GSD (Getting Stuff Done), helps create intensity by sharing updates and setting goals. (41m32s)
• Six-week reviews also contribute to creating intensity by allowing teams to reflect on their progress and make adjustments. (41m37s)
• Metrics and alerts are used to monitor the system and identify potential issues, enabling teams to take corrective action and maintain intensity. (41m40s)
• The concept of "code yellow" is used to signal that a particular area, such as the developer experience, needs attention and resources to be allocated to resolve the issue. (42m2s)
• Building infrastructure layers, such as the one created to improve the developer experience, allows teams to move faster in the long run, even if it takes longer to build initially. (42m32s)
• An example of building an infrastructure layer is the creation of a platform that exposes APIs, enabling anyone to build features like NFT gating in a short amount of time. (43m25s)
• This approach, referred to as "putting gas in the tank," allows teams to build a foundation that can be used to drive future development and innovation. (43m45s)
• The key consideration in building infrastructure layers is to think about what can be built on top of the platform, rather than just focusing on a single use case. (43m53s)
• A key aspect of building a high-intensity culture involves thinking about software in a different way, focusing on building infrastructure layers that may take longer to develop initially but allow for quicker development of features on top of it (44m0s).
• This approach involves writing code against APIs that don't exist yet, refining the client and server, and then implementing the server code, which can take two to three months but ultimately enables faster development of features (44m14s).
• The goal is to build a platform that allows others to build on top of it quickly, rather than focusing on building the best possible point solution (45m8s).
• This approach also emphasizes the importance of crafting great code and pair programming, as the platform will be used by others to build on (45m18s).
• The company has a focus on deleting unnecessary code and simplifying its codebase, with the goal of making it easier to maintain and build on (45m45s).
• GitHub co-pilot has written over one million lines of code for Shopify, but the company's goal is to have co-pilot delete one million lines of code, indicating a focus on simplicity and elegance (45m37s).
• The company has deleted millions of lines of code this year as part of its efforts to focus on rebuilding things elegantly and eliminating unnecessary code (45m55s).
• The company's goal is to have a smaller codebase, with the CEO tweeting that he wants Shopify to be ranked 100th in the list of top Ruby codebases, rather than competing for the top spot (46m14s).
• Simplifying a code base is crucial for improving resiliency, performance, reliability, and maintainability, as it makes everything easier to manage and maintain, even in a large and complicated set of tools like Shopify (47m23s).
• To achieve this, developers need to have a clear understanding of the code base and the domain expertise to identify areas that can be simplified or deleted (47m46s).
• Shopify has a "delete code club" and hack days, where teams focus on deleting unnecessary code, and they have a manual to help find things to delete (48m3s).
• These efforts have resulted in the deletion of over a million lines of code, which is a significant achievement, but also highlights the need for more efficient coding practices (48m21s).
• The goal is not to write more code, but to write elegant and efficient code, and to think about deleting unnecessary code, rather than just focusing on writing new code (48m50s).
• Shopify prioritizes long-term infrastructure thinking and cares about the elegance and simplicity of its code base, recognizing that the right code is not about the volume of lines, but about its quality and efficiency (49m0s).

Three buckets of building (49m5s)

• Building a high-intensity culture involves prioritizing speed, velocity, and intensity, which can be achieved by having a smaller and cleaner codebase that allows for faster movement and is more durable, although this can be challenging for companies to prioritize (49m6s).
• To address this challenge, the company thinks about new builds in terms of three buckets: experiments, features, and infrastructure, which helps determine the approach and resources needed for each project (49m42s).
• Experiments are used to learn and may turn into features or infrastructure, while features take advantage of existing infrastructure, and infrastructure builds enable multiple use cases and can take longer to develop (49m50s).
• Getting into the right mindset for this approach can be difficult, but having a leader like Toby who pushes the team to think about infrastructure and simplicity can make a significant difference (50m35s).
• Toby's approach involves encouraging the team to generate more options and not settle for the first two or three ideas, but instead to explore the 10,000 possible right options and find the optimal one (51m0s).
• This approach has changed the way the company works, with team members learning to spot patterns and push for simplicity, deletion of unnecessary code, and infrastructure development (51m36s).
• The benefits of this approach include free performance, easy navigation, maintainability, and resiliency, which are achieved through the hard work of deleting unnecessary code and building infrastructure (51m48s).
• Choosing the hard thing and making features easy to build is crucial for creating a high-intensity culture (51m59s).
• Pair programming is beneficial as it eliminates multitasking, allowing individuals to focus on the task at hand, and when done correctly, can make the pair of programmers feel like a genius (52m20s).
• The concept of pair programming is inspired by the idea that "alone we're okay, but together we're a genius," a quote from Michael Lewis' book "The Undoing Project" (52m56s).
• Demo culture is another key aspect of creating intensity, where high-fidelity updates are shared through demos rather than just screenshots, allowing for a more immersive experience and quicker identification of issues (53m20s).
• Shopify uses a tool called Spin, an internal development environment, to share demos and encourage feedback, which helps to short-circuit misunderstandings and create intensity around the fidelity of feedback (53m41s).
• The use of beta flags and friction logs also facilitates high-fidelity feedback, enabling teams to quickly identify and address issues, and iterate on the experience (54m16s).
• Shopify has around 11,000 employees and is valued highly, with a strong operating rhythm that keeps people moving and focused on the right tools and goals, providing quick feedback if they're moving in the wrong direction (55m3s).
• The company's operating rhythm involves regular check-ins that pair leaders with employees to work on problems together, combining the expertise of both the leaders and the employees, known as "Crafters" (55m32s).
• Strategic meetings and check-ins are valued at Shopify, with the goal of blocking obstacles and making progress, rather than simply eliminating meetings (55m58s).
• The engineering environment at Shopify is designed to be a "crafter's paradise," where engineers and other technical employees can practice their craft and improve their skills, using tools like Tuple for pair programming (56m15s).
• The company aims to provide a great experience for employees in various roles, including engineering, UX, PM, and ML, and wants to be the best place for them to work (56m32s).
• Teams at Shopify are set up for success by minimizing distractions, using tools like pair programming and a shift from Slack, and having a flexible approach to working hours (56m51s).
• The company has a six-week cycle of check-ins and reviews, which helps employees stay focused and motivated, and provides an opportunity to adjust their approach if needed (57m15s).
• Shopify is a remote company, with employees all over the world, but still maintains a strong sense of collaboration and teamwork (57m42s).

Remote work and trust battery (57m45s)

• The company has a 90-95% remote work setup, with intentional in-person experiences to foster trust and connection among employees, including an annual event called Shopify Summit that brings the whole company together for talks, hack days, and social activities (57m45s).
• In addition to Shopify Summit, the company has "bursts" where employees can gather in person to work on specific problems or projects, allowing for in-person collaboration and trust-building (58m22s).
• The company also has offices that employees can use if they want, allowing for random interactions and a sense of connection (58m51s).
• The main reason for the remote work setup is to be able to hire the best people in the world, regardless of their location (59m11s).
• The company uses the "trust battery" metaphor to describe the level of trust between employees, which can deplete over time if not nurtured through in-person interactions (58m42s).
• The trust battery can be charged through intentional in-person experiences, and it's up to each individual to manage their own trust battery and use it strategically (59m59s).
• Different managers use the trust battery metaphor in different ways, such as starting new employees at 50% trust or assuming high trust from the beginning (1h0m9s).
• The trust battery terminology is used as a shortcut to figure out how to work with someone and to navigate relationships within the company (1h0m25s).

Finding stability in uncomfortable times (1h0m29s)

• The personality of the leadership team is quite intense, with many founders on the executive team who are naturally impatient and intense people by design, and even non-founders are accomplished individuals with a similar attitude of impatience (1h0m53s).
• The leadership team tries to maintain a high level of energy and progress by sharing updates and showing progress on various projects, which creates a notion that there's always a lot going on and encourages others to jump in and contribute (1h1m26s).
• Having a high-energy and intense team is reminiscent of the culture at Airbnb, where no matter how well things were going, there was always a sense of needing to go faster and do more (1h1m44s).
• It's essential to have two or three big projects that are on fire at any given time, as this indicates that the team is pushing hard enough and taking necessary risks, and if everything is going well, it may be a sign that the team is not trying hard enough (1h2m0s).
• Having a few things on fire at all times is not about creating chaos but rather about stretching into new things, going faster than expected, or having new leaders who are being counted on early, which creates a sense of uncertainty but also drives growth (1h2m17s).
• Embracing a little bit of chaos at the edge is necessary for a business to become incredibly successful, and not having this chaos can be even more stressful and painful in the long run (1h2m45s).
• The idea that comfort gives stability is incorrect, and instead, it's the uncomfortable situations that provide stability because they encourage constant learning and make the team more robust against challenges (1h2m59s).

Hiring philosophy (1h3m14s)

• Interviews are not a good predictor of performance, as studies have shown that people who interview well may not perform well in the job, and vice versa (1h3m33s).
• A better approach is to use a "job trial" or "pair programming" to assess a candidate's skills and fit with the company culture, as seen in the example of two machine learning engineers hired at a startup (1h4m1s).
• One candidate, a PhD holder, was let go within a few weeks due to a poor cultural fit, while the other candidate, who had no prior software experience but was enthusiastic about machine learning, became a phenomenal engineer (1h4m11s).
• The key to successful hiring is to avoid bias and give candidates a chance to work in the company environment, using methods like pair programming to assess their skills (1h4m28s).
• The "race car analogy" is used to illustrate the idea that the best way to assess a candidate's skills is to put them in a real-world situation, rather than relying on interviews (1h5m8s).
• Shopify spends time in the 30-60-90 days after hiring to ensure that the candidate's skills align with the company's needs, and is transparent with candidates about the company culture and expectations (1h5m21s).
• If a candidate is not a good fit, it is in the best interest of both the candidate and the company to figure this out quickly, so the candidate can find a better fit elsewhere (1h5m32s).
• Job trials and intern programs are seen as effective ways to assess a candidate's skills and fit with the company culture, as they provide real work product and a chance to work in the company environment (1h6m1s).
• Shopify's hiring process involves a four-month trial period to assess a candidate's fit for the company, allowing both parties to determine if it's a good match before committing to a full-time position (1h6m13s).
• This approach is preferred over traditional interviews, as it provides a more accurate representation of a candidate's skills and work style, resulting in lower attrition rates (1h6m40s).
• The company's hiring process includes technical evaluations and a "life story" interview step, which aims to understand a candidate's experiences, motivations, and curiosity (1h7m41s).
• The "life story" interview is designed to go beyond a candidate's resume and understand the reasons behind their career choices and decisions (1h7m44s).
• The goal is to identify individuals who are curious, have range, and can adapt to different roles and challenges, as these traits are valued at Shopify (1h7m54s).
• The company's approach to hiring is centered around creating an interview process that closely resembles the actual job, allowing for a more accurate assessment of a candidate's skills (1h7m27s).
• Shopify's compensation system was redesigned by the VP and Head of Engineering, who, despite being an engineer, took on the task, exemplifying the company's emphasis on generalists and adaptability (1h8m52s).
• The book "Range" by David Epstein is mentioned as an influence on the company's approach to hiring and valuing generalists over specialists (1h8m34s).
• Shopify has a survey for new hires to gauge their happiness and fit within the company, which is used to provide feedback and adjust expectations, and if necessary, find a different role within the company or elsewhere (1h9m1s).
• The goal is for employees to have high impact, whether at Shopify or another company, and to ensure that expectations are set and met (1h9m30s).
• Shopify does not currently do work trials with new engineering hires due to the high volume of resumes, but they do have a large internship program (1h9m40s).
• The internship program is used as a form of interview process, where interns work on useful projects and can be considered for full-time positions (1h10m13s).
• Shopify plans to hire 1,000 interns in 2025, which will provide a large pool of candidates to choose from for full-time positions (1h9m51s).
• The internship program is seen as a win-win-win, as interns bring new skills and perspectives, and can either stay at Shopify or take their skills to another company (1h10m59s).
• The program is also seen as a way to restart the early talent ecosystem, which has been impacted by layoffs and other challenges in the industry (1h11m16s).
• Other companies have taken notice of Shopify's internship program and are planning to hire more interns as well, which is seen as a positive development for the industry (1h11m6s).
• Interns are often more intense than full-time employees, which can help drive the company's culture of intensity (1h11m31s).

Internship programs and co-op systems (1h11m41s)

• The internship program at Shopify emerged from the Co-op system that Canadian schools have, with the goal of providing early talent with a cohort and in-person experience (1h11m41s).
• Shopify's intern program requires interns to come in three days a week, rather than being fully remote, and is currently only available in three offices: Montreal, Toronto, and Ottawa (1h12m35s).
• The program aims to provide interns with a cohort of dozens of people, as well as regular in-person interactions with their mentors or managers, who are required to meet with them at least once a month (1h13m14s).
• The internship program is designed to give early talent a better experience, with the option to become full-time employees and work remotely if desired (1h13m34s).
• Co-op programs, like the one at the University of Waterloo, allow students to alternate between school and work terms, providing them with two years of experience by the time they graduate (1h13m41s).
• The Co-op program at Waterloo involves alternating between school and work terms every four months, allowing students to gain intense work experience and apply what they've learned in school (1h13m50s).
• Many schools, including those in the US, offer similar Co-op programs, and they can be a valuable way for employers to find and hire talented students (1h14m26s).
• The Co-op program can also be a symbiotic relationship between schools and employers, providing benefits to both parties (1h14m21s).
• Some companies, like the speaker's startup, have successfully hired and retained Co-op students, who can bring valuable skills and experience to the company (1h14m40s).

Lessons from managing 120 direct reports (1h15m32s)

• Extreme Labs, a company that emerged from Canada's tech scene, had a unique culture and work environment, with the goal of creating a "Crafters Paradise" where people didn't need managers to solve problems (1h15m37s).
• The company started small, with 10 people, and grew to have 120 direct reports under one person, which was an experiment to see if managers were necessary for solving problems (1h16m7s).
• To avoid needing managers, the company implemented various solutions, such as pair programming, product backlogs, demos, and set working hours, which helped to unblock employees and provide feedback (1h16m48s).
• The company's approach to management involved having a lot of direct reports, but not doing scheduled one-on-ones, instead opting for unscheduled one-on-ones, which allowed for more flexibility and clarity (1h17m22s).
• The unscheduled one-on-ones allowed employees to come and ask questions, and the manager could unblock them, providing clarity and solutions to problems (1h18m6s).
• The company's work environment was intense, with set working hours, and the manager was always available to answer questions and provide feedback (1h17m6s).
• The approach to management was an experiment, and the results showed that it was possible to create a work environment where managers were not necessary for solving problems, but rather for providing clarity and solutions (1h18m13s).
• At Shopify, a high-intensity culture was built by creating a flat organizational structure, where one person had 120 direct reports, and later, a few people were made directors with 40 direct reports each, to maintain a flat hierarchy and allow for the use of systems to unblock people rather than relying on one-on-one meetings (1h18m21s).
• The flat organizational structure was created to avoid having a deep hierarchy, where three people report to someone who only has three people reporting to them, and to keep things as flat as possible, with a recommended range of 8 to 20 direct reports per person in engineering (1h20m2s).
• The flat organizational structure is believed to be beneficial, as it allows for better alignment and reduces the distance between employees and the CEO, Toby, with survey results showing that the farther employees are from Toby, the more alignment gets out of whack (1h20m20s).
• Chamath, a major investor, played a role in changing the organizational structure by pointing out that the current system may not work at 400 people, leading to the implementation of a more structured system while maintaining a flat hierarchy (1h18m55s).
• The use of systems to unblock people, rather than relying on one-on-one meetings, was a key aspect of the flat organizational structure, allowing for more efficient use of time and resources (1h19m30s).
• Experimentation and trying new approaches, rather than following traditional methods, was a key aspect of building the high-intensity culture at Shopify (1h19m49s).

Failure corner (1h20m40s)

• Creating a "failure resume" by writing down past failures can be a depressing experience, as learned from attempting to follow Tim Ferriss' advice, and is not recommended (1h21m15s).
• Being laid off twice was a significant failure, but it allowed for reevaluation and growth, and was ultimately the right decision for both the company and personal development (1h21m31s).
• A major failure occurred in 2019 when rebuilding Shopify's Point of Sales system, where a hedged solution was chosen to build the iOS version in Swift and the Android version in React Native, resulting in wasted time and resources (1h21m53s).
• The decision to build the Android version in React Native was based on a desire to learn about the platform and the assumption that it would be easier to develop, but ultimately proved to be a mistake (1h22m21s).
• A year after launching the successful iOS version, the Android version was launched, but it was quickly realized that React Native was the superior platform, allowing for a single platform to run on the web and utilizing React Engineers from the web (1h22m42s).
• The mistake resulted in 18 months of wasted time and resources, involving around 100 engineers, and required rewriting and rebasing the code onto the iOS version (1h23m21s).
• The experience taught a valuable lesson, and the mistake was acknowledged and addressed by going back to the team and explaining the error, with the support of Toby (1h23m31s).
• A decision was made to not take a risk on a platform that had not been launched in production, which resulted in 18 months of wasted time for 100 engineers, and this decision is considered a total failure (1h23m34s).
• The mistake made was not taking a risk, and it was emphasized that not taking risks is often viewed more harshly than taking a risk that doesn't work out (1h24m0s).
• The correct approach would have been to take the risk on the unlaunched platform, as it would have saved time in the long run (1h24m26s).
• The company is now successfully using React Native, and they are core contributors to the project, with even the Shop app being built on React Native (1h24m34s).
• The decision to not take the risk was likely due to not considering the full extent of the potential outcomes and not being informed enough to make the right choice (1h25m7s).
• In hindsight, it's clear that the riskier path would have been the correct choice, and this experience serves as a valuable lesson in the importance of taking calculated risks (1h25m12s).
• The individual involved in the decision has a tendency to fully commit to a project or technology, as seen in their involvement with React Native, where they became a core contributor and release captain (1h25m21s).
• This experience highlights the importance of considering one's own thought process and potential outcomes when making decisions, and not hedging incorrectly due to a lack of consideration (1h25m47s).
• A decision was made to hedge on a path, which is considered the worst option, and the CTO at the time considered forcing a switch to React Native, but ultimately didn't make the decision (1h26m19s).
• Facebook had a similar experience early on, where their market cap dropped 80% due to difficulties with mobile ads, setting them back a year and a half (1h26m41s).
• Mark Zuckerberg discussed this experience in an interview at the Chase Center, stating that despite the setback, it wasn't too bad in hindsight (1h26m47s).
• The speaker was at Facebook, working on the Facebook app in extreme Labs, and experienced the challenges of submitting the iOS app, which kept getting rejected and crashing (1h27m8s).
• The team had direct access to people at Apple, but still struggled to ship a stable app, with new versions crashing every day (1h27m25s).
• The speaker was present during the HTML 5 fiasco, but did not advise Facebook to use HTML 5, they were just there when it happened (1h27m36s).

Lightning round and closing thoughts (1h27m46s)

• A performance management framework from a First Round Review article has been used for years, with its effectiveness attributed to the framework itself rather than the person using it, as it pulls out good information and has resulted in positive feedback from multiple people in the review process (1h28m5s).
• The framework was originally found through a Google search and has been used for every review, with the person considering using LLMs to make writing reviews easier (1h28m26s).
• The framework's origin is attributed to Vlad Lozovsky, a former manager at Airbnb, who may have invented it or been inspired by someone else (1h29m32s).
• Another framework, the W framework, has also been written about in a First Round Review post and is used by many companies for planning, often without knowing its origin (1h29m51s).
• A recommended book is "Mana" by Marshall Brain, which explores a future where AI tells humans what to do, and is considered an interesting read, especially in the context of AI development (1h30m36s).
• The book "Business Adventures" by John Brooks is recommended, as it provides in-depth analysis of problems, requiring focus to fully understand and learn from it (1h31m8s).
• Another recommended book is "Manna" which is an easy read, contrasting with "Business Adventures" (1h32m5s).
• The post "How to Do Great Work" by Paul Graham is also mentioned, highlighting its length and the importance of maintaining focus to extract valuable information from it (1h31m22s).
• A favorite recent movie is "Challengers", a tennis movie with a unique cinematography and music style (1h32m16s).
• The TV show "Halt and Catch Fire" is highly recommended, as it accurately portrays the early tech startup scene (1h32m42s).
• The Meta Ray-Ban sunglasses are a favorite product, offering a convenient and unintrusive way to listen to music, take calls, and record videos (1h33m4s).
• A favorite life motto is "everything you know is wrong," which encourages thinking from first principles and questioning established knowledge, even if it seems correct (1h34m13s).
• This motto promotes experimentation and trying new things, even if they may seem incorrect or unconventional, as seen in the example of switching gears in a Tesla without fully stopping the car (1h34m58s).
• The idea that "everything you know is wrong" is also related to the concept that everything around us was designed by someone, and it's possible that someone else could find a better solution (1h35m46s).
• Steve Jobs had a similar idea, stating that everything can be designed by people, and it's possible to create something new and better (1h35m57s).
• A story is shared about hiring a waitress who was observed doing an exceptional job at a restaurant, demonstrating strong organizational skills and attention to detail, and was offered a job at Extreme Labs (1h36m36s).
• Farhan Thawar shares a story about a woman he met at a restaurant, Extreme Labs, who initially worked as a receptionist and later became a recruiter and eventually took over one of the HR functions, and with the experience she gained, she was able to finish her university degree and become a director of HR at another company (1h37m3s).
• The woman was organized, smart, and had great people skills, but lacked experience with tools like Google Docs, which Farhan had to teach her (1h37m25s).
• Farhan also hired other people from the same restaurant who were very good at their jobs and had intense personalities, and many of them ended up on amazing career paths (1h38m1s).
• Farhan believes that everyone is smart at something and tries to figure out what people are good at and how he can work with them (1h38m31s).
• Farhan can be found online on Twitter, where he is active and tries to hang out the most, and his handle is @FN (1h38m49s).
• Farhan is open to being challenged and learning from others, and would love to hear from people who disagree with his views or have research that contradicts his opinions (1h39m0s).
• Listeners can be useful to Farhan by commenting and sharing their thoughts on what he got wrong, and he is open to learning from others and improving his ways (1h39m16s).