Engineering

Scaling from 50 to 100 employees almost killed our company. Until we discovered a simple org structure that unlocked $100M+ in annual revenue. In my 10+ years of experience as a founder, one of the biggest challenges I faced in scaling was bridging the organizational gap between startup and enterprise. We hit that wall at around 100~ employees. What worked beautifully with a small team suddenly became our biggest obstacle to growth. The problem was our functional org structure: Engineers reporting to engineering, product to product, business to business. This created a complex dependency web: • Planning took weeks • No clear ownership  • Business threw Jira tickets over the fence and prayed for them to get completed • Engineers didn’t understand priorities and worked on problems that didn’t align with customer needs That was when I studied Amazon's Single-Threaded Owner (STO) model, in which dedicated GMs run independent business units with their own cross-functional teams and manage P&L It looked great for Amazon's scale but felt impossible for growing companies like ours. These 2 critical barriers made it impractical for our scale: 1. Engineering Squad Requirements: True STO demands complete engineering teams (including managers) reporting to a single owner. At our size, we couldn't justify full engineering squads for each business unit. To make it work, we would have to quadruple our engineering headcount. 2. P&L Owner Complexity: STO leaders need unicorn-level skills: deep business acumen and P&L management experience. Not only are these leaders rare and expensive, but requiring all these skills in one person would have limited our talent pool and slowed our ability to launch new initiatives. What we needed was a model that captured STO's focus and accountability but worked for our size and growth needs. That's when we created Mission-Aligned Teams (MATs), a hybrid model that changed our execution (for good) Key principles: • Each team owns a specific mission (e.g., improving customer service, optimizing payment flow) • Teams are cross-functional and self-sufficient,  • Leaders can be anyone (engineer, PM, marketer) who's good at execution • People still report functionally for career development • Leaders focus on execution, not people management The results exceeded our highest expectations: New MAT leads launched new products, each generating $5-10M in revenue within a year with under 10 person teams. Planning became streamlined. Ownership became clear. But it's NOT for everyone (like STO wasn’t for us) If you're under 50 people, the overhead probably isn't worth it. If you're Amazon-scale, pure STO might be better. MAT works best in the messy middle: when you're too big for everyone to be in one room but too small for a full enterprise structure. image courtesy of Manu Cornet ------ If you liked this, follow me Henry Shi as I share insights from my journey of building and scaling a  $1B/year business.

Exciting updates on Project GR00T! We discover a systematic way to scale up robot data, tackling the most painful pain point in robotics. The idea is simple: human collects demonstration on a real robot, and we multiply that data 1000x or more in simulation. Let’s break it down: 1. We use Apple Vision Pro (yes!!) to give the human operator first person control of the humanoid. Vision Pro parses human hand pose and retargets the motion to the robot hand, all in real time. From the human’s point of view, they are immersed in another body like the Avatar. Teleoperation is slow and time-consuming, but we can afford to collect a small amount of data.  2. We use RoboCasa, a generative simulation framework, to multiply the demonstration data by varying the visual appearance and layout of the environment. In Jensen’s keynote video below, the humanoid is now placing the cup in hundreds of kitchens with a huge diversity of textures, furniture, and object placement. We only have 1 physical kitchen at the GEAR Lab in NVIDIA HQ, but we can conjure up infinite ones in simulation. 3. Finally, we apply MimicGen, a technique to multiply the above data even more by varying the *motion* of the robot. MimicGen generates vast number of new action trajectories based on the original human data, and filters out failed ones (e.g. those that drop the cup) to form a much larger dataset. To sum up, given 1 human trajectory with Vision Pro  -> RoboCasa produces N (varying visuals)  -> MimicGen further augments to NxM (varying motions). This is the way to trade compute for expensive human data by GPU-accelerated simulation. A while ago, I mentioned that teleoperation is fundamentally not scalable, because we are always limited by 24 hrs/robot/day in the world of atoms. Our new GR00T synthetic data pipeline breaks this barrier in the world of bits. Scaling has been so much fun for LLMs, and it's finally our turn to have fun in robotics! We are creating tools to enable everyone in the ecosystem to scale up with us: - RoboCasa: our generative simulation framework (Yuke Zhu). It's fully open-source! Here you go: https://robocasa.ai - MimicGen: our generative action framework (Ajay Mandlekar). The code is open-source for robot arms, but we will have another version for humanoid and 5-finger hands: https://lnkd.in/gsRArQXy - We are building a state-of-the-art Apple Vision Pro -> humanoid robot "Avatar" stack. Xiaolong Wang group’s open-source libraries laid the foundation: https://lnkd.in/gUYye7yt - Watch Jensen's keynote yesterday. He cannot hide his excitement about Project GR00T and robot foundation models! https://lnkd.in/g3hZteCG Finally, GEAR lab is hiring! We want the best roboticists in the world to join us on this moon-landing mission to solve physical AGI: https://lnkd.in/gTancpNK

Last week, I described four design patterns for AI agentic workflows that I believe will drive significant progress: Reflection, Tool use, Planning and Multi-agent collaboration. Instead of having an LLM generate its final output directly, an agentic workflow prompts the LLM multiple times, giving it opportunities to build step by step to higher-quality output. Here, I'd like to discuss Reflection. It's relatively quick to implement, and I've seen it lead to surprising performance gains. You may have had the experience of prompting ChatGPT/Claude/Gemini, receiving unsatisfactory output, delivering critical feedback to help the LLM improve its response, and then getting a better response. What if you automate the step of delivering critical feedback, so the model automatically criticizes its own output and improves its response? This is the crux of Reflection. Take the task of asking an LLM to write code. We can prompt it to generate the desired code directly to carry out some task X. Then, we can prompt it to reflect on its own output, perhaps as follows: Here’s code intended for task X: [previously generated code] Check the code carefully for correctness, style, and efficiency, and give constructive criticism for how to improve it. Sometimes this causes the LLM to spot problems and come up with constructive suggestions. Next, we can prompt the LLM with context including (i) the previously generated code and (ii) the constructive feedback, and ask it to use the feedback to rewrite the code. This can lead to a better response. Repeating the criticism/rewrite process might yield further improvements. This self-reflection process allows the LLM to spot gaps and improve its output on a variety of tasks including producing code, writing text, and answering questions. And we can go beyond self-reflection by giving the LLM tools that help evaluate its output; for example, running its code through a few unit tests to check whether it generates correct results on test cases or searching the web to double-check text output. Then it can reflect on any errors it found and come up with ideas for improvement. Further, we can implement Reflection using a multi-agent framework. I've found it convenient to create two agents, one prompted to generate good outputs and the other prompted to give constructive criticism of the first agent's output. The resulting discussion between the two agents leads to improved responses. Reflection is a relatively basic type of agentic workflow, but I've been delighted by how much it improved my applications’ results. If you’re interested in learning more about reflection, I recommend: - Self-Refine: Iterative Refinement with Self-Feedback, by Madaan et al. (2023) - Reflexion: Language Agents with Verbal Reinforcement Learning, by Shinn et al. (2023) - CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing, by Gou et al. (2024) [Original text: https://lnkd.in/g4bTuWtU ]

Demystifying the Software Testing 1️⃣ 𝗙𝘂𝗻𝗰𝘁𝗶𝗼𝗻𝗮𝗹 𝗧𝗲𝘀𝘁𝗶𝗻𝗴: 𝗧𝗵𝗲 𝗕𝗮𝘀𝗶𝗰𝘀: Unit Testing: Isolating individual code units to ensure they work as expected. Think of it as testing each brick before building a wall. Integration Testing: Verifying how different modules work together. Imagine testing how the bricks fit into the wall. System Testing: Putting it all together, ensuring the entire system functions as designed. Now, test the whole building for stability and functionality. Acceptance Testing: The final hurdle! Here, users or stakeholders confirm the software meets their needs. Think of it as the grand opening ceremony for your building. 2️⃣ 𝗡𝗼𝗻-𝗙𝘂𝗻𝗰𝘁𝗶𝗼𝗻𝗮𝗹 𝗧𝗲𝘀𝘁𝗶𝗻𝗴: 𝗕𝗲𝘆𝗼𝗻𝗱 𝘁𝗵𝗲 𝗕𝗮𝘀𝗶𝗰𝘀: ️ Performance Testing: Assessing speed, responsiveness, and scalability under different loads. Imagine testing how many people your building can safely accommodate. Security Testing: Identifying and mitigating vulnerabilities to protect against cyberattacks. Think of it as installing security systems and testing their effectiveness. Usability Testing: Evaluating how easy and intuitive the software is to use. Imagine testing how user-friendly your building is for navigation and accessibility. 3️⃣ 𝗢𝘁𝗵𝗲𝗿 𝗧𝗲𝘀𝘁𝗶𝗻𝗴 𝗔𝘃𝗲𝗻𝘂𝗲𝘀: 𝗧𝗵𝗲 𝗦𝗽𝗲𝗰𝗶𝗮𝗹𝗶𝘇𝗲𝗱 𝗖𝗿𝗲𝘄: Regression Testing: Ensuring new changes haven't broken existing functionality. Imagine checking your building for cracks after renovations. Smoke Testing: A quick sanity check to ensure basic functionality before further testing. Think of turning on the lights and checking for basic systems functionality before a deeper inspection. Exploratory Testing: Unstructured, creative testing to uncover unexpected issues. Imagine a detective searching for hidden clues in your building. Have I overlooked anything? Please share your thoughts—your insights are priceless to me.

Drone shows are increasingly incorporating AI technologies to enhance their performance. What do you think about this one? Here are several ways in which #AI is being utilized in drone shows: 1. Autonomous Navigation: Path Planning: AI algorithms assist drones in planning and optimizing flight paths for intricate aerial displays. Collision Avoidance: AI enables real-time analysis of the environment, helping drones avoid collisions and maintain safe distances. 2. Formation Flying: Coordination Algorithms: AI algorithms coordinate the movements of multiple drones to achieve precise formations. Real-Time Adjustments: Drones can dynamically adjust their positions in response to environmental factors or unexpected changes. 3. Swarm Intelligence: Collective Behavior: AI-driven swarm intelligence allows drones to exhibit collective behavior, creating synchronized and mesmerizing patterns. Adaptability: Drones in a swarm can adapt their behavior based on the actions of neighboring drones. 4. Real-Time Data Analysis: Environmental Sensors: Drones equipped with sensors provide real-time data on weather conditions, wind speed, and other factors. Adjusting Performances: AI analyzes this data to make real-time adjustments to the drone show, ensuring optimal performance. 5. Light and Color Choreography: Dynamic Lighting: AI algorithms control the lighting elements on drones, creating dynamic and customizable light shows. Color Synchronization: Drones can synchronize their colors and lighting patterns in real time for visually stunning effects. 6. AI-Generated Patterns: Generative Algorithms: AI is used to generate unique and artistic patterns for drone formations. Variability: Each show can be different, adding an element of surprise and creativity. 7. Gesture Recognition: Audience Interaction: AI-powered gesture recognition systems allow drones to respond to audience movements or gestures. Interactive Shows: Audience members can influence the show in real time. 8. Dynamic Choreography: Learning Algorithms: AI can learn from previous performances, adjusting choreography based on audience reactions and preferences. Continuous Improvement: Drones can adapt and improve their performances over time. 9. Logistics Optimization: Efficient Deployment: AI assists in optimizing the deployment and retrieval of drones before and after shows. Battery Management: Algorithms manage drone battery usage for extended performances. 10. Safety Measures: Emergency Protocols: AI can implement emergency protocols to ensure the safety of the drone show, such as automated landing in case of malfunctions. Monitoring Systems: AI monitors drones for any irregularities in flight behavior. 11. Sound Integration: Audio-Synchronized Displays: AI synchronizes drone movements with music or other audio elements for a fully immersive experience. #ai #innovation via @ zzmenx #drone #dronetechnology

In the last 24 months we identified 300+ new legislations related to climate change and over 10% of them have elements assessing green claims. But what are the steps for a business to comply with the upcoming legislation in the EU? To comply with the EU's greenwashing regulations and avoid misleading consumers, companies should take the following steps: 1. Review and audit all marketing materials and environmental claims: Businesses should conduct a thorough review of their marketing materials and environmental claims to ensure they align with the regulations. This may involve consulting with legal and sustainability experts to identify potential areas of concern. 2. Substantiate environmental claims: Companies must provide evidence to support their environmental claims, using credible and verifiable sources. This may include scientific studies, third-party certifications, or government data. Companies should be prepared to disclose this information if required by the regulations. 3. Rigorous carbon accounting:  To prove one’s environmental impact, you will have to back it up with data. Companies must diverge from industry averages when calculating the footprint of a product or service. It is important to leverage primary activity data with already existing proof, for example, your scope 1 and 2 can be easily tracked through energy invoices, bills and such. Then, the golden share still is represented from scope 3 emissions, but it is important for companies to start backing up their claims with proof and data. 4. Implement standardised environmental labels: The EU Commission promotes using standardised environmental labels, such as the EU Ecolabel, to provide consumers with reliable information about a product's environmental performance. Companies should consider adopting these labels where applicable to demonstrate compliance with the regulations. 5. Train employees on greenwashing and regulations: Companies should provide training to their employees on greenwashing to ensure that all relevant personnel understand the implications of these regulations and can identify potential compliance issues. 6. Continuously monitor and update marketing materials: Businesses should regularly review and update their marketing materials and environmental claims to ensure ongoing compliance with regulations. This may involve keeping abreast of new developments in sustainability research, as well as changes to the regulatory environment. To understand further how the EU greenwashing regulations will impact your business, have a read here: https://lnkd.in/egrfuk6h To understand green-related terms, have a read here: https://lnkd.in/eznWaTZ5 #greenwashing #sustainability #co2 #eu #co2 #esg #compliance

This is probably the most valuable tip I share with students and clients who want to get ahead in their professional lives: → Track your wins!! In a document (Excel, Word, or whatever works for you), create three columns: 1. TASK – What was it? ↳ Led a team meeting to resolve a bottleneck in the project timeline. 2. ACTION – What did you actually do? ↳ Facilitated a structured discussion to identify roadblocks, proposed a revised workflow, and reassigned tasks based on individual strengths and deadlines. 3. IMPACT – What measurable difference did it make? ↳ Reduced project timeline by 15%, increased task completion rate by 20%, and improved overall team alignment and morale. Update it at the end of each week. It’s such a simple approach, but it ensures you’re always ready to showcase your value when it matters most - whether it’s for performance reviews, job interviews, or pitching yourself for your next big opportunity. Highly recommend it! P.S. Have you ever tried something like this to keep track of your achievements? #careergrowth

#Diversity in high-tech fields remains critically low. The Equal Employment Opportunity Commission (EEOC) recently reported that #Black and #Latino professionals are underrepresented in high-tech roles, especially in leadership. These numbers highlight ongoing structural barriers in hiring, promotion and retention. This gap is a missed opportunity to tap into a wealth of diverse talent and perspectives essential to the future of tech. However, addressing and thoroughly fixing these challenges will require time, consistent effort and a long-term commitment to systemic change. Companies can support the progression of representation in tech by investing in training, mentorship and internship opportunities that open doors for people who were historically shut out. Programs like internXL, a platform that is committed to increasing diversity and inclusion in the internship hiring process for top companies, are making a significant impact. Similarly, the expansion of STEM education at institutions like Cornell University is helping to connect talented young people from underrepresented communities with opportunities for high-tech careers. When we work together to remove these barriers, we’re fostering a more inclusive workforce and strengthening innovation, problem-solving and leadership in the industry. Let’s build a tech future that reflects the diversity of our society. https://bit.ly/3UNtOCh