Josefa Paz España Herrera

What No One Tells You About Being a Senior Software Engineer: — You’re no longer hired to write code—you’re hired to make your team’s code better. Your impact goes far beyond your commits. — Teaching is your superpower. The faster you can level up junior engineers, the more valuable you become. — Your technical opinions will shape products. But with great influence comes the responsibility to back opinions with data. — Soft skills matter more than hard skills. Communication, empathy, and influence make the difference between good and great engineers. — You’re not just solving problems—you’re anticipating them. Proactively identifying risks and gaps is your job now. — You need to unblock others. A senior engineer who can keep the team moving is worth their weight in gold. — Architectural decisions are expensive. Make them with care and consider future maintainability. — Delegation is a skill. Learn to let go and trust others—you can’t do everything yourself anymore. — Your feedback defines the culture. The way you review code or guide a discussion sets the tone for the team. — You’re not just building features—you’re building systems that scale. Think beyond the current sprint and plan for the long term. If You’re: • An aspiring tech lead looking to grow influence • A senior engineer aiming for staff or principal roles • A technical expert ready to shape engineering culture …these truths will guide your journey. Seniority isn’t just a title—it’s a mindset shift.

819

31 comentarios

Bad estimates kill more software projects than bad code. Treat them like deadlines and you guarantee late nights, broken features, and pissed-off teams. Here’s the truth: - Complexity multiplies cost. - Unknowns always show up. - “Easy” work is never easy. When I’m asked for one, I follow 7 rules: 1. I remind them it’s a guess, not a promise. 2. I assume complexity will cost more. 3. I expect unknowns to show up. 4. I treat “easy” tasks as traps. 5. I give ranges, not dates. 6. I call out my assumptions. 7. I update estimates as the system evolves. Amateurs chase certainty. Pros manage risk. That’s how you deliver without breaking your team. Add more 👇

301

85 comentarios

As a SOFTWARE Engineer with 6+ years of experience working with global companies - these are the BEST pieces of advice I’ve ever received 🏆 1️⃣ Write code for humans, not just machines The best code isn’t the most complex - it’s the one that’s easy to read, maintain, and scale. 2️⃣ Your network is as important as your skillset The best opportunities don’t always come from job boards. They come from conversations. 3️⃣ Don’t chase the new shiny tech - master the fundamentals first Frameworks change. Languages evolve. But strong problem-solving skills will always be relevant. 4️⃣ Feel free to ask better questions The smartest engineers aren’t the ones with all the answers. They’re the ones who ask the right questions. 5️⃣ Soft skills > Hard skills (in the long run) You can be a coding wizard, but if you can’t communicate your ideas or collaborate, your growth will hit a ceiling. 6️⃣ Know when to walk away That bug you’ve been stuck on for hours? Step away. Take a break. The answer will often reveal itself. What’s the best advice you’ve ever received in your career? ♻️ If you found this helpful with your network!

733

68 comentarios

"Software Engineering is dead there is no more job" Last 24 hours there were more than 8000 new jobs on Linkedin in USA. When you filter further for keywords like junior or senior you will that there are ~1600 jobs for Senior Engineers and ~50 for Juniors. Other roles were for other advanced engineering roles that require 3-10 years of experience. The industry is thriving yet people on reddit and twitter are calling for armageddon. You just got to skill-up and learn how to build software. Don't get intimidated by the keywords and requirement for multi-years it's usually just a filter for 'we want someone who isn't a noob' Senior is just a keyword for someone who understand what he is doing. Honestly i saw people in senior engineering roles that the interns run circle around. This is also fair, you live in the age of LLMs. 2018 grad students only had to show calculators and todo apps because they actually built those app themselves. Copying stack overflow doesn't really work you gotta understand the code and know how to apply it in the context of what you are building. In 2025 you can generate 2018 graduate output in less than 10 minute with a code agent and you are expecting to be measured with the same standard ? You want 6 figures for writing React in 2025? Are you are wrong - why would someone pay you huge sums of money when LLMs can do frontend development better than you (check v0) There are those who understood this and pivoted to becoming a product engineer which incorporate other skills that are valuable to businesses or descended to lower layers of the stack that can't be automated by AI like QA, DevOps, Cloud (Reliability will always have a lot of people working on it) Conditions have changed but the effort required has remained the same. You can do more now so more is expected of you.

430

130 comentarios

I just read "Hire and Train Haskell Junior Developers" by Elvie Ciobanu, and it resonated with my experience. My experience hiring junior developers who didn't even know Haskell coming in and building a system with them showed me that two common statements are clearly mistaken: - it's hard finding developers to work on Haskell - using "fancy" extensions or libraries in Haskell makes it harder to recruit and more uncertain that everything will go according to plan I found out that although teaching myself Haskell had been pretty difficult, and I know one other very smart senior developer who faced similar difficulties, mentoring junior devs to help them learn Haskell was pretty easy. Finding junior devs willing to learn Haskell on the job wasn't difficult either. And I totally agree with her that it's how we mentor them and how we communicate with each other, not only with the junior devs but as a team, that makes it possible to use the fancy parts of Haskell without them making it harder to work on the code (which they can, for sure). It's nothing revolutionary: many people are eager to learn, communicate well, create a safe environment for learning and cooperation, and it will be easy to grow a great team that creates great software.

39

2 comentarios

I completely reject the premise that developers need to learn to use LLMs in order to stay relevant. LLM-based coding as an inevitability is a lie pushed by companies that stand to make (or lose) a whole lot of money depending on if this is a bubble or not.

379

198 comentarios