Phi Nguyen

If you're a Senior Engineer looking to move up, the next role will likely be a Staff Engineer. And in this guide, Shruti shares tips from her own experience of getting promoted to Staff Engineer at PayPal and Slack. She talks about what Staff Engineers do (and how it's different from Seniors), why you might not be getting promoted, and how to take that next step. https://lnkd.in/g58dnFEG

103

3 Comments

If you're a Senior Engineer looking to move up, the next role will likely be a Staff Engineer. And in this guide, Shruti shares tips from her own experience of getting promoted to Staff Engineer at PayPal and Slack. She talks about what Staff Engineers do (and how it's different from Seniors), why you might not be getting promoted, and how to take that next step. https://lnkd.in/g58dnFEG

80

Former GitHub CEO - Thomas Dohmke just came out of stealth with Entire : first open-source CLI called Checkpoints that captures agent context alongside code every time you commit Thomas Dohmke, former GitHub CEO, raise $60M in seed funding for his new startup Entire — now valued at $300M — to rethink how we build software in the age of AI agents. This isn’t just another dev tool — Entire is launching its first open-source CLI called Checkpoints that captures agent context alongside code every time you commit. That means the why, how, and what behind AI-generated code won’t be lost in chat logs or ephemeral sessions anymore. Why this matters for the AI-driven future of software development: Context as first-class data: Entire CLI hooks into your Git workflow and stores session history (prompts, responses, files changed, reasoning) as versioned checkpoints alongside code — making agent work traceable, reviewable, and auditable. Bridging human + agent workflows: Traditional tools assume humans write code — but with AI agents writing more every day, we need tooling that treats agent reasoning as part of the development lifecycle. Entire is setting that foundation. Open ecosystem & transparency: By launching this as open source, Entire is giving the community a way to build standards for agent context tracking rather than locking this capability in proprietary platforms. This is a huge milestone in the evolution of developer experiences — turning AI agent output from opaque artifact into traceable, collaborative process that humans can understand and improve. If you’re working with AI assistants, multi-agent systems, or just thinking about how to scale developer productivity in 2026 and beyond, this is worth a close look. https://lnkd.in/gZ8rEm9s #AI #DeveloperTools #OpenSource #SoftwareEngineering #CLI #AIagents #SoftwareContext

1

why limit yourself to someone else's preference and curation? go build a db go build a web framework go write a file system go build an operating system go build a prog language go write a webserver go build an ORM go build an agent build whatever excites you you have all the resources in the world to build literally anything you want. why follow a roadmap!

1,161

43 Comments

I have spent the last decade arguing with leaders, managers, peers, and juniors about how important maintaining a high quality codebase is for driving feature work forward. [cue my past coworkers rolling their eyes because I've actually been really annoying about it]. I have cared so much about creating systems that allow product engineers to write high quality code efficiently that I've dedicated my career to it. I have been told many times "if you want to get promoted, focus on being a product engineer, not an infrastructure engineer." But AI is causing these norms to shift, from my vantage point. It feels as if overnight, one's ability to write, and one's ability to understand and apply business context has gone from severely undervalued to crucial. Now, companies are rushing to build guardrails into their systems to prevent AI from going off the rails. Now, suddenly, test coverage, types, and CI systems are considered important. Google's new AI guidance to software engineers emphasizes “'maintaining rigor' across code review, security, and maintenance." It makes me wonder why we're willing to focus on building high quality systems when it helps the machines, but not when it helps the people. https://lnkd.in/eyKhEZx3

119

8 Comments

🦺 🛟 Agent Safety > Model Safety LLMs didn't break your app. Agents can. As we wire LLMs to tools, browsers, payments & code the risk shifts from bad text to bad actions. Top failure modes (seen in the wild & in papers): 🚩 Prompt/Content Injection -> Tool Misuse. Malicious web pages or data instruct agents to steal secrets or make unintended API calls. This is now a primary risk for browsing and MCP - style tool agents. 🚩 Over permissioned tools & long-lived tokens. Agents get "god mode" scopes. A single injection becomes account takeover or irreversible ops. Security notes for MCP emphasise least privilege and short-lived auth. 🚩 Unsafe web autonomy. Benchmarks show web agents will attempt harmful tasks unless constrained (posting misinformation, illicit sales, etc.). You must measure misuse not just assume guardrails. 🚩 Supply-chain & retrieval poisoning. Agents trust plugins, third-party tools & indexed data that an attacker can taint. OWASP’s newest GenAI Top 10 calls this out explicitly. 🚩 Process safety gaps. NIST's GenAI Profile (AI 600-1) warns that organizations ship agents without role clarity, human-in-the-loop or incident playbooks. ✅ Model safety reduces bad text. Agent safety prevents bad state. Treat the agent like a junior SRE - with narrow permissions, audited actions and clear escalation paths. If you're building agents, what's the one control you won't ship without? Also drop a comment if you'd like me to share a follow-up post on how to tackle each of these risks. #AgenticAI #AISafety #AIAgents #GenAI #CyberSecurity #MCP #OWASP #NIST #AITrust #ResponsibleAI #AIForBusiness

5

I’ve been writing 99% of my code at Airbnb with LLMs (mostly GPT-5 w/ Codex). If you think this is a bold claim it means you’re doing it wrong 😅 When you're building high quality production software, writing code is always the 𝗹𝗮𝘀𝘁 𝘀𝘁𝗲𝗽. Your first step is to understand the problem that needs to be solved. Then ideate solutions, consider alternatives, explore tradeoffs and refine your exploration into a concrete plan. Even as you implement the plan task by task you should not be coding a stream of conscious. That leads to bad code design. You should be considering the architecture of the code, abstractions and coming up with a clean way to write it. Only after all this upfront design and planning work do you then start manually typing code with your fingers. That last step is not necessary to do manually anymore. Whenever I think of coding, I immediately reach for an LLM because I use it like a power tool. A carpenter does not leave their power drill on the table when they need to screw in a bolt. Why would you not use an LLM to execute on your plan? This is not Vibe Coding, you are not relying on the LLM to decide on its own how to implement your solution. You are in the driver's seat, providing direct technical guidance at every step. 𝗬𝗼𝘂𝗿 𝗲𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲 𝗮𝗻𝗱 𝘀𝗸𝗶𝗹𝗹 𝗹𝗲𝘃𝗲𝗹 𝗱𝗶𝗿𝗲𝗰𝘁𝗹𝘆 𝗶𝗺𝗽𝗮𝗰𝘁 𝗵𝗼𝘄 𝗴𝗼𝗼𝗱 𝘁𝗵𝗲 𝘀𝗼𝗹𝘂𝘁𝗶𝗼𝗻 𝗶𝘀. No, this is not slower than doing it without LLMs. You should also use LLMs as power tools for research, planning and architecture. This will get you even higher quality software than without them. It allows you to go far beyond due diligence and truly explore, analyze and refine your design fully before any single line of code is written. I use the following workflow to naturally research, design and plan the feature I want to build in the form of a conversation which then gets converted to a formal Spec that the LLM can implement task by task:  1. Explain the problem to the LLM.  2. Give it your ideas for the initial solution  3. Tell it explicitly: “Propose an approach first. Show alternatives to my solution, highlight tradeoffs. Do not write code until I approve.”  4. Review the proposal, poke holes in it, iterate  5. Tell it to write the plan to disk as a spec so you can hand off to another session later  6. Lastly, let it generate code.

1,462

186 Comments

Game changer: Claude Code paired with Playwright CLI Playwright allows computers to work with the visual hierarchy of web pages, interact with page elements and take screenshots. Other ways to interact with web pages from Claude include the Claude Chrome extension and the Playwright MCP server. Both options are heavy on token use because they force the LLM to process images and a large MCP surface area. With the Playwright CLI you get token efficient browser processing. By default it's headless but you can also have it show what is seen in a window. Some ways Claude + Playwright CLI helped me recently. Testing Some AI generated code was completely broken on some API calls. The problem was the data shape was wrong so the tests "passing" meant nothing. Adding Playwright in the loop provided Claude a way to test the code in the same way a user would use the app. Iterative Design I pointed Claude at the current state and the reference for a "good" design I wanted to move towards. I interacted in real time regarding how the page was looking and directed test cases to further check behavior. I am currently exploring other use cases and I'm excited for a functional level up in my workflow. Are you doing visual work with AI? What tools are you using?

5

{{ Titles Don’t Define Your Worth in Tech }} In the tech world, titles really don’t matter as much as we think. You could be a Tech Lead in one company, then move to another company with a higher package but be called a Senior Engineer. Why? Because roles and titles depend on context: 👉 In your current company, your skills may be enough to lead a team. 👉 In the new company, the tech stack might span multiple domains, and your expertise fits best in a focused role. The takeaway? Embrace knowledge, not titles. Growth comes from what you learn, not what’s written on your business card. 💡 Focus on continuous learning and impact, the titles will follow. #SoftwareEngineering #Leadership #GrowthMindset #CareerGrowth #SkillsOverTitles #ContinuousLearning

4

2 Comments

You’ll never become a great software engineer if you’ve never maintained production services. Shipping features is easy.  But maintaining them in production, where customers actually use and break your services, is where you grow. Being a great engineer isn’t about how much code you write.  Or how many features you ship.  Or what you can deliver in a sprint. It’s about ownership across the entire SDLC, not just the “fun” parts. This includes oncall, planning, hiring, mentoring, stakeholder management and much more. Because once your code hits production, speed alone isn’t enough. → Customers expect reliability. Your service must work every time. → Teams expect judgement. You need to make sound engineering decisions. → Systems demand resilience. Real-world failures are guaranteed. And with the rise of AI-assisted vibe coders, this is even more important. Anyone can ask Claude or Gemini to generate code. However, owning that change across the entire SDLC is a completely different ball game. If you want to grow as an engineer, focus on end-to-end ownership: → Understanding customer behaviour → Communicating trade-offs → Balancing speed vs reliability → Adding features without breaking stability → Managing technical debt → Performing migrations safely → Learning from incidents → Improving developer experience → Tightening performance and reliability → Owning operational processes → Redesigning systems when assumptions break This is the skill set that turns good engineers into great ones. And it’s how you build systems customers can trust. #softwareengineering

728

64 Comments