Did some playing around with n8n telegram, webhooks, bots. End result is that I create a daily reminder that sends out an AI Generated message to my telegram account and I can check and interact with it on my phone. Telegram integration is so much easier than Slack.

A recent CBS News piece highlights a big shift in how companies are thinking about talent. Instead of focusing mainly on degrees or job titles, many employers are prioritizing practical, demonstrable skills. According to LinkedIn career expert Catherine Fisher, the ability to work confidently with AI tools is quickly becoming one of the most valuable capabilities a candidate can showcase. As more organizations move toward skills-based hiring, showing real familiarity with AI can make applicants stand out in a crowded job market. Interesting read on how AI literacy is moving from “nice to have” to “must have” for many roles: https://lnkd.in/ei4XcsNe

n8n vs Antigravity Hot take correction ... I may have been wrong about n8n. I’ve been pretty vocal about my skepticism of low-code orchestration tools like n8n. My main concern was always control: API flexibility, real logic, and whether you could build serious agentic workflows without fighting the platform. So… I decided to test that belief. I took one of my real projects and rebuilt it using Google’s Antigravity instead of n8n ... same objective, same constraints. Result? I was wrong about what this class of tooling can actually do now. Using Antigravity, I was able to: Hit the USAJobs API cleanly Run structured job searches (Tampa Bay, Data Analyst / Power BI / SQL roles) Score matches against my resume Automatically generate and send custom job-match emails Produce a refined, role-specific cover letter without manual glue code Here’s an example of the email the agent generated and sent on a 76% match: Subject: 🚀 USAJobs High Match: Program Analyst (76%) Agency: Department of Homeland Security Location: Tampa Bay Followed by a tailored cover letter that actually reflects my background in Power BI, SQL, automation, and data-driven decision making. No brittle hacks. No “workflow spaghetti.” Just a clean agent loop: observe → score → act. I still think n8n has limitations but I’ll own this: agent-first platforms have crossed a threshold, and dismissing them outright is no longer a serious position. Lesson learned: Don’t argue tools in theory. Port a real project and let reality decide. If you’re experimenting with agentic workflows, job search automation, or resume-aware scoring systems, I’m happy to compare notes. http://aiforhire.net

Hey everyone, I am building a RAG for a lawyer company (working with ton of documents). What do you think is the best way to build RAG (Lang Chain, Lang Graph, Supabase, etc.). Also ideally to be bit more private, as we are working with sensitive information. Any advice will be appreciated!

Can someone help me make an ai receptionist for a salon that can book, reschedule, or cancel appointments. I need the ai to sound realistic and remember client base.