Category: Try

Date: 02/25/2025

Watch the Video

Okay, this video on implementing long-term memory in AI agents using n8n is seriously inspiring. It walks you through building an AI assistant that remembers user preferences, just like ChatGPT. Forget stateless interactions – we’re talking persistent memory that makes your AI feel way more human.

Why is this gold for us developers diving into the AI/no-code space? Because it bridges the gap between traditional coding and LLM workflows. We’re used to managing state in our apps, but now we can offload that to a no-code platform like n8n *and* leverage LLMs to make sense of that state! The video shows you exactly how to save memories to Airtable and retrieve them to inform future interactions. Think about the possibilities: personalized customer support, dynamically tailored learning experiences, or even AI-driven project management that actually *learns* from past projects.

I am pretty excited to try this out. I can see myself using something like this to automate client onboarding, and I can’t wait to explore how this type of memory could drastically improve the user experience in our products. It’s not just about building cool AI toys, but about creating genuinely useful, adaptive applications.

Date: 02/25/2025

Watch the Video

This video on self-hosting Lovable with Coolify is super relevant for anyone, like myself, who’s diving into the world of AI-powered development and automation. It essentially tackles a common problem: how do you scale and control your deployments as your AI-driven projects grow, without breaking the bank with managed services? The video walks you through setting up Coolify, connecting it to your GitHub repos (including private ones!), and automating deployments with webhooks. Plus, it even covers troubleshooting common issues like bad gateway errors, which, let’s be honest, we’ve all been there!

What makes this video particularly valuable is that it demonstrates how to leverage no-code/low-code tools like Coolify to handle the deployment pipeline. This frees up time to focus on the core AI coding aspects of a project. Imagine using AI code generation to rapidly prototype a new feature, then having Coolify automatically deploy it to a self-hosted environment. Also, the promise of integrating self-hosted Supabase in part two, now that’s interesting, because that means you are in full control of your data too!

From my perspective, as someone who’s always looking for ways to streamline my workflow and reduce reliance on external services, this video is a goldmine. The ability to self-host and automate deployments allows for greater flexibility and cost control. And the peace of mind in knowing that you control your own servers, your models and your databases is a fantastic proposition. I’m definitely experimenting with Coolify this weekend!

Date: 02/23/2025

Watch the Video

This FlowiseAI intro video is pure gold for anyone, like me, diving headfirst into AI-powered workflows. It’s a walkthrough on setting up Flowise, an open-source, low-code platform. The video covers three different ways to set up Flowise: local installation, cloud deployment, and Flowise Cloud.

Why is this valuable? Because it bridges the gap between traditional coding and the power of AI. We can leverage visual interfaces to build AI solutions without getting bogged down in complex code, which accelerates development and opens doors to more creative AI integrations. Think about automating customer support with AI-driven chatbots, or even streamlining internal data processing with custom AI workflows.

For example, I’m currently working on a project where we’re using LLMs to analyze customer feedback. Traditionally, this would involve writing a ton of custom code. But with Flowise, I can visually connect different AI components, like language models and data connectors, to create a streamlined process. The fact that the video explores deployment options – local, cloud, and Flowise Cloud – is super practical. It gives you the flexibility to choose the setup that best fits your needs and resources. I’m particularly interested in Flowise Cloud to save myself the headache of managing infrastructure. It’s definitely worth experimenting with to see how it can speed up your AI projects and free you up to focus on the creative side of things.

Date: 02/23/2025

Watch the Video

This video on Supabase’s Row Level Security (RLS) is gold for anyone like me who’s been diving headfirst into AI-assisted development. It basically shows you how to enforce data access rules *right in the database* itself using PostgreSQL’s RLS policies, rather than relying solely on your application code. It walks you through using the Supabase dashboard to visually set up policies that control who can see, edit, or delete data, and shows how these rules play out across different tables and user roles. It even demos using Supabase’s AI tools to simplify policy creation.

What makes this particularly valuable is that RLS can become an essential part of your application’s security architecture. It’s all about moving security closer to the data itself. Instead of relying on potentially buggy or leaky API code to filter data, you define the rules at the database level. This means queries sent directly to the database are automatically filtered based on the user’s role and permissions. The video clearly explains how you can test these policies to ensure they work as expected for different user types. It’s a massive shift towards more robust and secure applications, especially as we start generating more code with AI. This also ties into the broader no-code/low-code movement, because Supabase AI tools are lowering the barrier to entry for complex security configurations, and they are doing it in a way that makes it auditable and repeatable in code!

Honestly, it’s worth experimenting with because it’s a fundamental piece of the puzzle when building secure, scalable applications. It’s no longer enough to just trust your API layer. With AI generating so much of our code, having that extra layer of database-level security gives you much more peace of mind. Plus, Supabase makes it surprisingly easy to get started, even if you’re not a database expert, especially using their AI tools. This is something I’m going to be incorporating into my next project!

Date: 02/22/2025

Watch the Video

Okay, this video looks like a goldmine for anyone, like me, who’s been diving headfirst into no-code automation! It’s all about connecting Salesforce with Google Calendar using n8n, which is a really cool, open-source alternative to Zapier. The video walks you through setting up workflows that automatically create and update Google Calendar events based on changes in Salesforce records. The best part is, it doesn’t just show you *what* to do, but *how* to do it, covering everything from setting up webhooks and outbound messages in Salesforce to using different nodes in n8n.

What makes this particularly valuable is how practical it is. As someone who’s spent years writing custom integrations, the idea of visually building these workflows with n8n is incredibly appealing. Imagine, instead of spending days coding and debugging API calls, you could achieve the same result with a drag-and-drop interface! Plus, the video also explains how to test these automations without racking up costs, which is a huge win. The video compares the approach to a similar one using Zapier, this kind of benchmarking is useful to show the trade offs in technology, in terms of cost and effort, that we need to make in evaluating these tools.

I’m itching to experiment with this! The outbound message approach, in particular, seems like a more robust and reliable trigger than some of the polling methods I’ve used in the past. Plus, I’m always looking for ways to reduce our reliance on expensive platforms and move towards more open and customizable solutions. Who knows, maybe n8n could become the new backbone of our integration strategy!

Date: 02/22/2025

Watch the Video

Okay, this video by Ahmed Mukhtar is seriously exciting, especially if you’re like me and trying to blend traditional development with AI. He walks through building a Voice AI Agent using Ultravox and N8N that’s both powerful *and* cost-effective. Think appointment scheduling, RAG, personalized interactions – the whole nine yards. But the cool part is how he’s optimizing the stack: ditching OpenAI’s RealTime API for Ultravox (open-source, yes!), swapping Replit for Railway to save a few bucks, and wrapping it all in FastAPI with Windsurf.

Why is this inspiring? Because it’s a concrete example of shifting from closed, often expensive, AI solutions to a more open, customizable, and budget-friendly approach. We’re talking about taking control of our AI workflows. Imagine using this to build a customer service bot that can understand complex queries, schedule appointments directly in your Laravel app, and all without breaking the bank. And the N8N backend? That’s where the real automation magic happens.

Honestly, the part that grabs me is the open-source angle and the focus on cost. We’ve all been there, lured in by the promise of AI only to be hit with a massive bill. The idea of building something this capable for 5 cents a minute? That’s a game-changer. I’m definitely going to be digging into Ultravox and N8N after this. It feels like a solid foundation for building more complex, AI-driven features into our existing projects. Worth a try? Absolutely.

Date: 02/21/2025

Watch the Video

Okay, so this NetworkChuck video is gold for us devs diving into the AI space. Essentially, it’s about automating web browser tasks using AI, showcasing a free, open-source alternative to OpenAI’s Operator. He walks through using Browser Use, an open-source project, to control a web browser with AI, potentially automating workflows.

Why is this valuable? Well, we’re moving beyond just writing code; we’re building systems where AI agents handle repetitive tasks. Think about automated testing, data scraping, or even filling out complex forms. The fact that it’s open-source and *free* means we can experiment without the $200/month Operator price tag. Being able to run this locally with tools like Ollama also means we can keep our data private and experiment without constant cloud dependencies.

Imagine integrating this into our Laravel applications! We could use it to automatically generate reports, monitor competitor pricing, or even handle customer support inquiries via a browser interface. For me, the real kicker is the potential for automating UI testing. Instead of writing countless Selenium scripts, we could teach an AI agent to navigate our app and identify issues. It’s absolutely worth experimenting with because it opens the door to building truly intelligent, self-operating web applications.

Date: 02/21/2025

Watch the Video

Okay, so Wes Roth’s latest video dives into the SWE-Lancer benchmark and OpenAI’s exploration of whether LLMs can actually *earn* money doing freelance software engineering. Seriously, can an LLM rake in a million bucks tackling real-world coding tasks? That’s the question!

This is gold for us as we’re moving towards AI-assisted development. Why? Because it’s not just about generating code snippets anymore; it’s about end-to-end problem-solving. The SWE-Lancer benchmark tests LLMs on real-world freelance gigs, meaning we can start to see where these models excel (and where they still fall short). This can directly inform how we integrate them into our Laravel workflows, maybe using them to automate bug fixes, generate boilerplate, or even handle entire feature implementations. The linked GitHub repo provides a tangible way to experiment with these concepts and see how they perform in our own environments.

For me, the potential here is huge. Imagine automating away those tedious tasks that eat up so much of our time, freeing us to focus on the higher-level architecture and creative problem-solving. This video isn’t just news; it’s a glimpse into a future where AI is a true partner in software development. Definitely worth checking out and experimenting with the benchmark. It’s time to see how we can leverage this stuff to build better apps, faster.

Date: 02/20/2025

Watch the Video

Okay, so this video is all about creating a completely automated, faceless YouTube channel using n8n. Forget manually scripting, filming, and editing – this walks you through setting up a workflow that uses tools like ChatGPT, ElevenLabs, Replicate, and Creatomate, all orchestrated within n8n. It’s basically a step-by-step guide to building a content creation machine.

Why is this gold for us devs exploring AI? Because it perfectly showcases how to string together different AI services into a cohesive, automated process. We’re talking LLMs for content ideation, voice synthesis for narration, image generation for visuals, and video editing APIs to compile everything. It’s a practical example of how to leverage no-code platforms like n8n to manage complex AI interactions without writing a ton of custom code.

Imagine applying this to other areas. Think automated report generation, marketing content creation, or even dynamic documentation. The power lies in understanding how to connect these AI building blocks within a workflow. I’m personally excited to try this out because it’s a tangible example of “AI coding” – less about writing algorithms from scratch and more about orchestrating existing AI services to solve real-world problems. Plus, who doesn’t want to see a fully automated YouTube channel in action? It’s worth the time investment to grasp the workflow design principles.

Date: 02/20/2025

Watch the Video

Okay, this video is all about Simon’s “Founder Stack,” a collection of software he uses to run his business, and it’s incredibly relevant to anyone diving into the AI-enhanced workflow. He showcases tools like Aidbase.ai and Feedhive.com, but also goes deeper into platforms like n8n.io, Replicate.com, and even ComfyUI for more advanced AI image generation. Plus, he mentions Cursor.com, which looks like a really interesting AI-powered code editor. He essentially presents a full ecosystem for automating tasks and leveraging AI across his business.

What’s inspiring here is the tangible application of these technologies. It’s not just theoretical hype; it’s a peek into how someone is *actually* using AI and no-code tools to build and manage a SaaS portfolio. For those of us transitioning from traditional PHP/Laravel development, it’s a goldmine of ideas. We can see how n8n.io could automate tasks we used to build from scratch, or how Replicate.com can integrate cutting-edge AI models directly into our applications without complex infrastructure setup. The inclusion of image generation hints at cool possibilities for dynamic content creation and personalized user experiences.

Honestly, seeing this makes me want to experiment with integrating ComfyUI or a similar solution into an application for handling complex image processing tasks that I previously would have had to write in PHP or Python. This is about shifting from “I can build that” to “How can AI help me build that *faster* and *better*?”. This video provides that inspiration and a concrete set of tools to start exploring.