About
Experienced Software Engineer building compelling user experiences with cutting edge web…
Activity
307 followers
-
-
Nathan Witt shared thisWASM blurring the line between native & web https://lnkd.in/eDJA3sZj
-
Nathan Witt shared thisWasn't expecting to see my name in Photoshop 2021's credits yet there it is, how cool! #adobelife
-
Nathan Witt shared thisI’m selling my 1 bedroom condo in Ukrainian Village. This is the BEST neighborhood I’ve lived in Chicago by far in my 15 years here, but we’re getting ready to relocate to California and now’s the time. Have a look and tell your friends! #forsale https://lnkd.in/ehRFV7j
-
Nathan Witt shared thisReady to find out where the next step in my career takes me! Looking for the right product company that knows how to build great software, especially if the problem domain relates to technology, aerospace, digital manufacturing and/or IoT. Currently living in Chicago, but also open to remote work, relocation to Austin, TX or the west coast. My contact information can be found at the bottom of my resume linked in this post
-
Nathan Witt liked thisNathan Witt liked thisI still marvel at the ingenuity of the New York programmer who modified an interest calculation routine to steal "round-off" pennies. He hid the money in an account named "Zzyzyx," knowing that auditors checked accounts alphabetically and would always stop before they reached the letter Z. It took years for a complete database audit to finally catch him. That level of planning is now obsolete. I recently went down a rabbit hole to see if I could bypass physical check stealing entirely. Sadly, within ten minutes and a few AI prompts, I had created a digital fake check image that mimics a real one perfectly. I didn't need to steal a check or engrave a printing plate; I just needed the account info and a prompt. The counterfeiter is no longer a skilled artisan; they are just a prompt engineer sitting in front of a computer. In this article, I explore the frightening ease of AI-assisted fraud—from fake invoices to land title theft—and why we must train our AI models not just to generate content, but to detect the very crimes they are now capable of committingFrom Zzyzyx to AI: The Modern Crime Wave No One is Ready ForFrom Zzyzyx to AI: The Modern Crime Wave No One is Ready For
-
Nathan Witt liked thisNathan Witt liked thisNew Behance project I did in collaboration with Rob Gregory Photography. Caitlin Clark is an explosive player who has taken over the WNBA (Women's National Basketball Association). Breaking records and leading her team to the playoffs for the first time in 7 years. Check out the full project: https://lnkd.in/emkSzm8Q #WNBA #Basketball #IndianaFever #CaitlinClark #C4d #DigitalArt
-
Nathan Witt liked thisNathan Witt liked thisIf you weren't on it last week, you're already late. Do better, copywriters.
-
Nathan Witt liked thisNathan Witt liked thisAppreciate the shout out (chart) to [esc] by a16z games .. the cat is peeking out of the bag. Facinating & relevant read. #escape, #neocinema, #ai, #unreal, #a16zThe Next Generation Pixar: How AI will Merge Film & Games | Andreessen HorowitzThe Next Generation Pixar: How AI will Merge Film & Games | Andreessen Horowitz
-
Nathan Witt reacted on thisNathan Witt reacted on thisHey friends! I just finished the "MVP" of my Product Portfolio website - the first iteration being just a fancier resumé. I'm already working on detailed case studies, but in the meantime, take a look and let me know what you think! https://www.mikelse.com/
-
Nathan Witt liked thisNathan Witt liked thisWhen you check out profiles on LinkedIn, do you browse publicly or invisibly? Why? For a long time, I was an anonymous lurker. I recently made the switch and...it's nice! Does it cause more people to look at my profile in return? For sure. But perhaps that's a good thing (it even got me a consulting project). If you see me, send a message! I'm (almost) always game to chat.
-
Nathan Witt liked thisNathan Witt liked thisWe partnered with Major League Baseball (MLB) ™/© to bring back fan kits for this 2024 baseball season. ⚾️ Share your love of the game with specialized GIFs and #AdobeExpress templates. Learn more: https://adobe.ly/3xwiGkr
Explore more posts
-
Deepthi Talasila
How GSD turns Claude into a self-steering developer The speed at which ClawdBot MoltBot OpenClaw climbed in popularity was quite phenomenal, and for good reason: It has an audience beyond the developer space, especially those who just want to experiment with digital assistance. Todays post continues my journey with GSD, but the LLM agent landscape is changing rapidly and this shooting star caught everyone’s eye. At its core, OpenClaw is a locally hosted, open-source AI assistant that runs on your computer or server. You talk to it through apps you probably already use, like WhatsApp, Telegram, Discord, Slack, Signal, or even iMessage. And it really “does stuff” — sometimes too well. It is a gateway service that maintains WebSocket connections and includes a smart orchestration layer for working with LLM agents. Stay connected for industry’s latest content – Follow Deepthi Talasila #DevSecOps #ApplicationSecurity #AgenticAI #CloudSecurity #CyberSecurity #AIinSecurity #SecureDevOps #AppSec #AIandSecurity #CloudComputing #SecurityEngineering #ZeroTrust #MLSecurity #AICompliance #SecurityAutomation #SecureCoding #linkedin #InfoSec #SecurityByDesign #AIThreatDetection #CloudNativeSecurity #ShiftLeftSecurity #SecureAI #AIinDevSecOps #SecurityOps #CyberResilience #DataSecurity #SecurityInnovation #SecurityArchitecture #TrustworthyAI #AIinCloudSecurity #NextGenSecurity https://lnkd.in/g8YPff7n
-
Huy Do
I maxed out my Supabase and Vercel free tiers. Instead of upgrading to a Pro plan, I built my own version of that. From scratch. 🤓 If you've ever bought a hosting package, you've probably seen that browser-based control panel where you set up your domain, create email accounts, manage databases. That's usually cPanel. With Claude Code I put together a full self-hosted control panel running on my own Hetzner server. 23 modules, one interface: ▸ Domains, DNS, SSL automated ▸ Database management with a built-in query editor ▸ Deployment pipeline (git pull, build, live) ▸ Live monitoring, logs, backups, scheduled tasks ▸ A full terminal running directly in the browser ▸ Multi-user access with role-based permissions ▸ Authentication with two-factor login (2FA) What made it actually work wasn't just the AI though. I set up shared files that multiple agents can read and write to, so they always know what's been done and what's next. The documentation gets updated in Notion automatically by Claude Code after every session. Same for the backlog: every feature listed in technical terms, plus a plain-English explanation 😅 so I always know exactly what was built and what's still missing. Not just "AI wrote my code" but a system where agents collaborate, document, and hand off context like a real team. And the thing I appreciate most: I now have one server that can run as many projects as I want. No per-project pricing, no usage limits, no platform deciding what I can or can't do. One fixed monthly cost, everything under my own control. Hitting that free tier limit might have been the best thing that happened to me. The foundation is set, the infrastructure is in place. From here I can just focus on building. Next up is still a native iOS app, so that one isn't going anywhere. 😉
-
Marcin Śpiewak
Figma Grid is here! (beta) It sounds great: organize, align, and place content into responsive layouts. But… It’s not really a full code like grid. It's more like a bento grid. 1. You can’t wrap items inside the grid. 2. No support for variables for gaps or sizes. 3. You can’t reuse column sizes as child sizes. 4. It is only responsive in one viewport at a time. So, if I want a 12-column grid on desktop and a 4-column one on mobile? Nope. If I want a 6-column child on desktop and full-width on mobile? Nope again. Config 2025 had a lot of cool stuff (Sites, Make, Draw with variables 😍), but GRID needs a lot more love. Please, Figma fix the grid. It could be amazing. 🙏
3 Comments -
Andy Poes
Incredible piece from Tom's Guide on Mixboard. Hype is real! "Google’s Mixboard lowers the barrier for those of us lacking graphic design skills. It's an AI tools that makes it easy to spark ideas with simple prompts and explore concepts in a fluid, middle-stage creative process." https://lnkd.in/gxWD5wKK
3 Comments -
MD Al Fariya Zisun
Ever feel like your website or app is missing that certain spark, that elusive 'je ne sais quoi'? Often, the answer lies not in flashy overhauls, but in the subtle power of ambient animations. I just read an insightful piece demonstrating how these passive, non-distracting movements can profoundly bring a design to life, enhancing brand identity, personality, and storytelling without overwhelming the user. It's about crafting an immersive atmosphere, a feeling that resonates, rather than just movement for movement's sake. From architecting robust Laravel backends to crafting dynamic user interfaces with React, or developing high-performance cross-platform applications with Flutter and React Native, my focus is always on delivering solutions that are not just functional, but truly captivating. Integrating thoughtful, performant, and accessible animation principles – whether through advanced tools like GSAP for complex morphing or lightweight CSS for subtle rotations – elevates a digital product from merely functional to truly engaging and memorable. It's about delivering a holistic experience that captivates your audience and reinforces your brand's unique narrative, all while prioritizing user experience and technical efficiency. What subtle design elements have you found most effective in creating a memorable and immersive user experience for your clients or users? #WebDevelopment #UIUX #Animation #DigitalExperience #TechConsulting #BangladeshTech
-
Ali Ali
There’s been a lot of talk lately about how “the design process is dead.” Jenny Wen from Anthropic mentioned it on Lenny’s podcast, and the conversation has been nonstop since. Personally, I don’t think the process is dead. It’s just changing. The traditional flow of discover, research, wireframe, mock, prototype, test wasn’t created because designers loved structure. It existed because building used to be expensive and understanding users took time. A wrong direction could cost real resources, so teams needed gates before committing to a build. That made sense at the time. But that process was built on a big assumption: that validating an idea had to happen before you built anything. That’s no longer true. With AI, small teams can move from idea to working product in a few days. You don’t always need months of research to understand an opportunity. You can build something real, get data, and actually feel how the experience works. That kind of learning is impossible from a Figma file or prototype. I still use Figma and my sketch book to explore visual directions or new interaction ideas. It’s how I think. But lately, I’ve been using AI tools to turn those explorations into real, data-driven prototypes I can validate right away. I get to test the flow, push on the edges, and feel where the design breaks down, sometimes before I even bring it to the team. For me, that’s where design is headed. Less about long, linear steps and more about cycling between exploration and real-world validation. The craft still matters. Taste still matters. Maybe even more now that building is faster and cheaper. The design process didn’t die. It’s shifting. And the best way to keep up is to stay close to the real experience, not just the idea of it.
4 Comments -
Mani Ramezan
The purpose of this post is to touch base on Ollama and how it can help with both privacy and running a cost efficient pipeline for indie devs. If you haven't heard of ollama, I highly recommend checking their website and read https://lnkd.in/gUvgcRSY. In a nutshell, it allows you to run certain models locally or in a docker container which can help with hosting models on your machine or server. It has a lot more features which I'm still learning about. I've been playing around with qwen3:30b for majority of the time and just recently updated to qwen3.5:35b. Both run smoothly on M4 Max / 48GB but could've run better on 64GB only if I knew better to get that upgrade. I'll write more on the projects that I'm working on in future, but in a nutshell, I used LLM to run a two-phase pipeline: - Filter 1.3million records from jsonl and reduce it to 50K useful records for the app - Run through all 50K records to select most meaningful samples among several examples for each record. In addition, fallback mechanism for any missing data needed
6 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content