🔥Telegram İfşa

505kg WORLD RECORD! History made again

Amusing use of LLMs at a more traditional company: “A project with ~50 people got stuck. There are too many JIRA tickets, no clear specification, and anytime one team tries to make progress, the others shoot it down. So a dev built an LLM to try and break the deadlock: (cont’d)

A friend asked me to look at their code, and *Holy Mother of God!* He works for a medium-sized company. They are building some sort of internal CRM. They've been vibe-coding some of the new features. They are professional developers. Most, I assume, know what they are doing,

Called it. Claude code is getting way stricter limits on the $200/month plan. No free lunch lasts forever.

A public company CEO told me AI coding has had negligible impact on his engineering teams, instead the real transformation has been on their product and design teams using Replit. I asked him how does he reconcile this with CEOs saying that 25-50% of code is generated by AI?

Migrated a 563 lines of raw prompt, 246 lines of LLM response processing, 48 lines of error handling that SRE team was using for ticket annotation to this 28 lines of simple DSPy beauty below ⬇️ Now going to extend the Annotate module to meet other functionality, but seeing

Perplexity is repeatedly modifying their user agent and changing IPs and ASNs to hide their crawling activity, in direct conflict with explicit no-crawl preferences expressed by websites. cfl.re/4l7RV9b

Some supposedly “reputable” AI companies act more like North Korean hackers. Time to name, shame, and hard block them.

🚀 Meet Qwen-Image — a 20B MMDiT model for next-gen text-to-image generation. Especially strong at creating stunning graphic posters with native text. Now open-source. 🔍 Key Highlights: 🔹 SOTA text rendering — rivals GPT-4o in English, best-in-class for Chinese 🔹 In-pixel

💙 We just love how excited TeamOhana Product Lead, Ravi Theja, was to talk about building Merit Letters into our Comp Planning module. 🎥 Watch the full webinar with Ravi and Tushar Makhija here: lnkd.in/gJ4nj_w8

techcrunch reports that AI coding tools have "very negative" gross margins. i.e. they're losing money on every user. this isn't just a temporary reality while competing in the land grab, it's the market telling us that this structure doesn't work. there's a natural market

Many thanks to Maxime Rivest 🧙‍♂️🦙🐧 for his advocacy in DSPy! Yesterday, thanks to his help in getting started, I contributed my first PR to improve structured outputs in DSPy (thanks to the amazing ideas put out there by Boundary ML (who built BAML). github.com/stanfordnlp/ds…

After programming for 16 years, I noticed that Software Engineer mentality slowly shifted from “produce great code” to “tolerate shit code”

Well, in the most lowkey way possible, and on a random Tue afternoon, DSPy 3.0 is out of beta. pip install -U dspy So many amazing people contributed to this. Thank you all! (Release notes below. Stay tuned for a ton of stuff over the next few weeks!!)

Going forward, AI assistance must be disclosed for Ghostty contributions: github.com/ghostty-org/gh…

📢 New on the blog: The FP&A gap that’s killing your headcount efficiency The FP&A gap costs companies more than wasted time. It’s costing them clarity, alignment, and confidence in their most important investment: their people. Read the article: teamohana.com/blog/fpna-gap

Adam Raine, 16, died from suicide in April after months on ChatGPT discussing plans to end his life. His parents have filed the first known case against OpenAI for wrongful death. My latest story on the psychological impacts of AI chatbots: nytimes.com/2025/08/26/tec…

Agentic AI is already in beta at TeamOhana and we've got big news coming very soon... Want to hear more about the future of workforce planning? Watch Tushar Makhija's interview with Joe Michalowski for the inside scoop on what we've been building ➡️ youtu.be/DWSmMTXr3UQ?fe…