Steven Harms

I’ve been building LeanEff, a small extensible effects library for Lean 4. LeanEff is based on Oleg Kiselyov’s work on freer and extensible effects, but shaped for practical Lean programming. It currently includes: - Concrete-row effect signatures - Reader, Writer, State, Except, Random, Clock, and IO-lifting effects - Handlers that remove effects from the row - Practical examples including ASCII Tetris, a SQLite-backed video rental shop, and a multi-agent snake arena with threaded and cooperative runtimes Repo: https://lnkd.in/eGdNXRhm

47

3 Comments

Here is Danielle Lindblom talking about making Web Accessibility a priority upfront to save on cost and effort. Gridiron Survivor team learning more about why a11y is important. Sponsored by: GitKraken Frontend Mentor Vercel

26

1 Comment

I didn’t think I needed another way to browse files. Turns out, I was wrong. broot is one of those tools that quietly fixes an annoyance you’ve learned to tolerate: navigating big, messy directories in the terminal. Instead of bouncing between ls, cd, and find, you get an interactive tree you can search, collapse, and jump through instantly. No setup drama. No new mental model. Just type, filter, move on. It doesn’t try to replace your shell: it just makes it feel smarter. If you work in large repos and live in the terminal, broot is a very easy win.

21

5 Comments

Anthropic restricted OpenClaw’s creator last week. To many, this looks like a platform or policy story. It’s not. It’s an architecture story. When your product logic is tightly coupled to a vendor SDK, an upstream change doesn’t just affect one feature — it ripples across your entire system: Pricing shifts. Access changes. Terms evolve. Suddenly, a working product becomes fragile. The instinct is to add another vendor. That’s not the fix: The real solution is designing a thin seam between your product and any external provider — so when something changes, you update one layer, not ten. But here’s the tradeoff nobody talks about: abstraction adds complexity, and sometimes limits how deeply you can leverage a platform. The question isn’t “which vendor do you choose?” It’s “how much of your product depends on them?” Because that’s what determines how fragile your system really is. If you're thinking about this right now, I broke down a simple way to audit your vendor surface area (takes ~7 minutes). Link in comments. 👇

5

1 Comment

I built a little something to make our lives easier. I've spent a lot of time thinking about why we still write so much boilerplate for APIs in 2026. Often, we find ourselves spending a lot of time, or recently LLM tokens, generating the same CRUD routes and serialisation logic repeatedly. As a developer, I just wanted to define my data and start building, without the "plumbing" or the prompts getting in the way. So, I built Crate - a D language framework that turns a simple struct API into a full REST, GraphQL, and MCP API in just three lines of code. Why is this different? While ecosystems like Rust (crudcrate), F# (Type Providers), or Haskell (Servant) have ways to reduce boilerplate, Crate is built on D's unique compile-time introspection. It doesn't just "generate" code probabilistically; it semantically understands your models to bake in authentication, pagination, and multi-protocol support automatically. It's been a labour of love to get that level of safety and flexibility working seamlessly. If you're a fan of clean code and saving time, I'd love for you to take a look. It's open-source and free to use: https://lnkd.in/dQ5jPerb. Feedback, stars, or just a "hello" are all very much appreciated! #OpenSource #SideProject #BuildInPublic #DLang #SoftwareEngineering #Rust #FSharp #WebDev #AI

16

SDK documentation is not supplementary material... it is a vital component of the SDK itself. It lets developers harness the SDK's potential by serving as a highly sophisticated technical bridge. However, the tone must be instructional, and the approach must be simple. Here is a quick explainer video on the need for SDK documentation and its essential components.

19

I believe this Layoff / AI funded world is different now - clean tickets are rare but impact is not optional (If your team needs step-by-step specs, you are managing tasks, not ownership) Matt Watson nails it - this is what product-driven leadership feels like. “Hand me a messy doc, unclear edges & a real user problem - I will ship something people remember ” 

2

Developers average 8–15 meetings per week. Many of those meetings shouldn't include them. Some shouldn't exist at all. But we accept by default-the invite arrives, we click yes, another hour disappears. Learning to decline meetings is a career skill. Here's what I tell the teams I work with: Decline status updates where you have nothing to add. Decline large meetings where your role is unclear. Decline recurring meetings that lost their purpose. Decline meetings that are actually email. The hardest one? Declining back-to-back chains that eliminate all focus time. Not because any single meeting is bad-but because the pattern destroys your ability to do actual work. Here's how to decline professionally: → Give a brief reason: "I'm protecting focus time for [project]" → Offer an alternative: "Can we move this to [specific time]?" → Suggest async: "Could you send this as a doc? I'll review and follow up" Most people think saying no to meetings will hurt their career. The opposite is true. Protecting your time to do high-impact work is what gets you promoted. Sitting in meetings where you add no value doesn't. What's one meeting on your calendar this week you should probably decline?

4

1 Comment

There's this chasm that appears right before/around Senior developer. If you got here through sheer effort and throughput, it is unlikely that more of the same will bridge the gap and continue your progression. We could talk multipliers and leadership and calculated risk, but there are so many ways to be a leader I prefer to start with reflection. What are your passions, values, and goals and how can those align to what your team needs and values? I love to start that discussion with a talk Sarah Stein gave last year about the path to Principal developer at Microsoft. At one point she talked about her 'Brand' as four bullet point statements that define her values and how she aims to be seen, and so the role she plays within her job. As an example, I'll share two of mine: - Zack has a viable solution to our biggest problems. (I aim to hold the bigger picture, bias toward action, and maintain an openness to alternative viable paths) - Zack is in this with us. (Career guidance, live site, priority negotiations, teaching/sharing, spec reviews -- I aim to be a collaborative and trustworthy peer always pushing toward the common goal) Anyone else have thoughts on their own Brand - those things you are or aim to be known for?

16

5 Comments

Sometimes it's worth stepping back and questioning the choices we made years ago - not because they were wrong, but to make sure they still make sense today. I spent some time diving back into CSS approaches recently, comparing plain CSS, CSS Modules, and various CSS-in-JS solutions. It was encouraging to see that Material UI has made real progress on accessibility features, though the performance overhead of CSS-in-JS remains a factor worth considering. This reminded me of something important: there's rarely one "best" solution that works for every project. The right choice depends on your team, your project's needs, and what you're trying to achieve. It's easy to get caught up chasing the latest framework or library, but sometimes the most valuable thing we can do is pause and ask: "What actually serves our users best?" The answer might surprise you. What's a technology choice you've revisited recently? Did it hold up to scrutiny? #CSS #CSSModules #CSSinJS #MaterialUI #WebDevelopment #FrontEndDevelopment

13

1 Comment

I was thinking about how I've coached new staff+ engineers to find and pitch impactful work, and the team at LeadDev helped polish those thoughts and get it published. A couple of places to start: * Where do we spend a lot of our time as an eng org? * What do we not do enough as an eng org? Asking these questions can point you to places where shifting the way of working can scale your impact a large part of the organization, without needing the constant, high-touch interactions we might have relied on when influencing just our team. (Link in the top comment because Feed Algorithms, I guess?)

27

1 Comment

Curiosity > Evaluation. There’s a pattern I keep noticing. We often spend more time trying to find what a person doesn’t know than understanding what they do know. Once you see it, it’s hard to ignore. - Meetings reward critique, not synthesis. - Interviews reward elimination, not discovery. - Social media rewards dunking, not depth A Better Way : In a professional setting , especially when you’re operating inside a team or building one, optimizing for signal beats gap-hunting on almost every axis that matters. Strong teams aren’t built by minimizing individual gaps. They’re built by recognizing signal: how someone thinks, what they’ve learned through experience, and where they consistently create impact. No one knows everything. High-performing teams succeed by stacking strengths, not by minimizing individual blind spots When I am participating a discussion (team meeting or interview) , My bias is simple: “optimize for signal, not for gotchas.”

3

You'd be surprised how many doors you can open just by Asking. An art that has to be learned extremely well is "Asking". A quick summary of what I learned by listening to a Jim Rohn's seminar is, 1) Asking is the beginning of Receiving. So, start the process. 2) Receiving is not the problem. Failure to Ask is. 3) Receiving is like an ocean. There's plenty. The problem is that some people go to ocean with a tea spoon. Two ways to Ask are: 1) Ask with Intelligence (Define & Describe clearly) 2) Ask with Faith (Believe you can get what you want like a child. Not like an Adult.) On a side note, a powerful phrase to start-over something is, "I'm Sorry". This has the power to reflect, explore, open a new pathway and fix broken windows. This has the power to clear most misunderstandings or conflicts in any relationship or matter. However, ensure to back it up with "Accountability" and always "Act" upon it with execution. #maturity #leadership #askhelp #saysorry

2

1 Comment

"Slop" is the new cool. I’m tired of software that feels like a weekend project masquerading as a pro tool. If you still value high-quality software, polished UX, and well-done truthful documentation, you’re in for a frustrating year. We are seeing a massive surge in "quick-and-dirty" shipping. Between AI-generated docs that say nothing and half-baked features rushed to market, it feels like mediocrity is being rebranded as "Agile." I just spent an hour wrestling with a new IDE designed for parallel AI sessions. The docs were useless, the UX was sloppy, and the build was so unstable I’m already retreating to my $n$-th VSCode fork. To top it off, we now have AI "assistants" embedded on every documentation site—hallucinating and lying 50% of the time, wasting my time and patience. Is this really the new standard?

39

2 Comments

With some time on my hands I've started looking at Astro to update my personal site and add a blog. It's Island Architecture documentation (https://lnkd.in/e46SeH2M ) caught my eye .. something seemed very familiar. I first implemented this approach working at an international classifieds site back in 2008. It's such a clean approach, I'm really glad it's been formalised and now being documented and used in frameworks. It's great to see an idea you liked validated by industry experts like Katie Sylor-Miller and Jason Miller ( https://lnkd.in/em7Pjs76 ). The problem that I had to solve, was the need to add dynamic paid advertising to pages without impacting the performance of heavily cached and largely static content, particularly front pages for popular categories, such as used cars or household goods. Building a section of the page in the browser, using pre-built HTML snippets fetched over AJAX allowed us to cache the entire page, and use CDN and local browser caching to avoid un-necessary requests to our very busy servers. (Back in 2007-2009, we were using beefy co-located servers, and any memory or other upgrades meant a trip to the datacenter, and rotating machines out of the cluster to be upgraded or replaced, rather than 10 minutes of clicking in AWS EC2 dashboards) The implementation was very straightforward - a hand-rolled vanilla javascript ajax fetch and DOM manipulation, simply appending the snippet to the scaffold element, no jQuery needed. The backend was an optimised mod_perl handler, that rotated through a cached list of ads to show, and incremented the count of views on each complete rotation, this reduced the database writes by between 20-fold and 100-fold, a key performance requirement when you have a cluster of replicated MySQL databases - read's are easy to scale, writes get tricky to scale very quickly, just ask Brad Fitzpatrick about how he found that with Livejournal.

7

1 Comment

🚧 API not ready yet? Don’t block frontend development. Extend it. I recently kicked off a greenfield React project using Relay, but the backend wasn’t fully built out yet. Rather than stub in brittle mocks or delay UI work, I leaned on Relay Client Schema Extensions to keep development moving. 🔧 I extended our GraphQL schema with mock data that gave us realistic, structured data we could query with actual Relay fragments 👥 The client schema became an API contract, making collaboration with our backend team more seamless 💡 Bonus: I used Cursor to generate Relay store update logic, saving time and boilerplate fatigue I wrote up about it here: https://lnkd.in/gGJ35-eS Curious if others are doing something similar. Have you used client schema extensions this way? How do you handle frontend momentum when the API is still a work in progress? #React #Relay #GraphQL #FrontEndDevelopment #TypeScript #DeveloperExperience #Cursor

5

You're sitting in a planning session, planning tasks, and then everyone checks how many credits they have left in Claude, and you only take into the sprint (which lasts a month) as much as fits in the tokens?

7

As my final act of vacation, I published the first article in a new series about the systems and processes I've been working on to make code example testing more accessible to the technical writers at my org. This entry point is an overview of the abstractions I designed to bring developer testing tools and process to a non-developer org. https://lnkd.in/eeXZx9XC #TechnicalWriting #DocsAsCode #CodeQuality

8