Ludi Li

Today we announced new capabilities for our Shop Direct and Buy for Me experiences on Amazon, enabling now feeds to facilitate external merchants to list reliably their product listings and reach millions of Amazon customers. Excited on this feature, that is a win/win experience for customers, merchants and brands. Excited to be part of such important project.

27

Optimizing OpenSearch: Shard Strategy and Best Practices Proper shard strategy is crucial for constructing high-performance, scalable, and dependable OpenSearch clusters. AWS OpenSearch Service offers guidance to assist architects in designing for efficiency and resilience. Key Takeaways from AWS Best Practices: - Right-Size Your Shards: Aim for 20–50 GB per shard initially to balance query performance, memory usage, and cluster management overhead. - Primary vs Replica Shards: Total shards = Primary shards × (1 + number of replicas). Replicas enhance read performance and fault tolerance. - Shard Count Considerations: Too many small shards lead to higher cluster overhead and slower recovery. Too few large shards result in slower queries and limited parallelism. - Plan for Growth: Anticipate future data volume, query patterns, and indexing frequency. Avoid default "one-size-fits-all" settings; customize based on workload. - Distribution Across Nodes: Distribute shards evenly across nodes to avoid hot spots. Monitor shard allocation and rebalance as needed. - Reindexing Strategy: Over time, adjust shard count or size through reindexing to uphold optimal cluster performance. Why This Matters: A well-considered sharding strategy ensures your OpenSearch cluster can manage high-volume, low-latency queries, scale effectively, and maintain resilience even during node failures. Reference: AWS OpenSearch Service – Shard Strategy Best Practices https://lnkd.in/gsrM6RSs

12

Early in my time at Amazon, I assumed documentation was mainly about transfer of knowledge. Later I realized it was also about preserving decisions against reinterpretation. In large organizations, memory is unstable. The team changes. The context shifts. A decision made for one reason gets explained later for a different one. The written record becomes more than a convenience. It becomes the only defense against architectural folklore. That is why I care less about document length and more about whether the tradeoff is captured clearly enough that six months from now nobody needs to guess what was actually decided.

56

3 Comments

I read an interesting piece about how Expedia Group is rolling out AI internally. Not the usual “AI will replace developers” story. Their CTO, Ramana Thumu, is doing something much simpler: They’re democratizing AI across the entire company. No central “AI team” deciding everything. Every team gets access. Every workflow can experiment. Every employee can build with it. They even built an internal AI playground where people can use multiple LLMs and create their own agents. More than 1,500 internal agents already built. That’s the interesting part. Not the technology. The operating model. Too many companies treat AI like a research project: • A central “AI initiative” • A small expert team • A roadmap disconnected from daily work But AI creates the most value when it lives inside real workflows. Inside engineering. Inside support. Inside operations. Where the actual problems are. The job of leadership is not to control AI. It’s to remove friction so teams can experiment safely. Because the biggest shift happening right now is simple: The companies that win with AI won’t have the best models. They’ll have the most people productively using them. #engineering #leadership #ai

6

As an SDM at Amazon, I often hear concerns that “𝗜𝗻𝘀𝗶𝘀𝘁 𝗼𝗻 𝘁𝗵𝗲 𝗛𝗶𝗴𝗵𝗲𝘀𝘁 𝗦𝘁𝗮𝗻𝗱𝗮𝗿𝗱𝘀” can become a bottleneck to delivery. That hasn’t been my experience. When practiced well, this principle isn’t about adding friction at the end of the process or holding teams to vague perfection. It’s about investing in a learning culture where engineers understand 𝘄𝗵𝘆 the bar exists and are empowered to raise it themselves through better design reviews, stronger testing, and clearer ownership. High standards, when paired with strong coaching and trust, actually accelerate delivery. Teams spend less time reworking issues, firefighting in production, or debating expectations. Instead, they build confidence in what they ship and continuously improve how they build it. For me, insisting on high standards is less about saying “no” and more about helping teams learn faster, make better trade-offs, and sustainably deliver great software at scale. That’s not a bottleneck but a leverage. The real question isn’t whether to insist on high standards, but how to practice them in a way that enables teams rather than slows them down. I’ll dive into that in my next post.

23

1 Comment

Someone lost their job at Amazon when they were expecting their first child and looking forward to being a parent. Now their whole world is upside down. Someone got laid off in October, fought hard to find an internal role, and still got impacted in the next round. Someone who spent 10 years in the company, worked across three teams, and managed people with care, also lost their job today. It happens quickly. You get an email, the next thing you know you are locked out of the system, badge stopped working, chats gone. It feels like you were never part of the place to begin with. Layoffs are the worst part of this industry, and you can do almost nothing to prevent them once a list is made. A few things I want to say to anyone who got impacted today: 1. A spreadsheet decision is not a measure of your intelligence, your kindness, or the value you created for users and teammates. Please be kind to yourself. 2. Take time to process. Cry, vent, sleep. You are a human who just had their safety rug pulled, not a robot who must instantly “bounce back.” 3. The projects you shipped, the skills you built, the people you helped, all of that is real. A layoff cuts pay, not your track record. Do not feel like you’re not strong enough to find your way again, you are. 4. Tell friends, ex colleagues, mentors what happened. Ask for referrals, mock interviews, resume reviews, even just calls where you talk about anything except work. If you were not impacted, check on the ones who were. Nobody chooses to be on the receiving end of that email. The least we can do is remind them that the email is not the end of their story.

972

37 Comments