Payal 🌻

@leojrr @leojrr payals-portfolio.vercel.app Vibe coded this! Please checkout

Meet the new Stitch, your vibe design partner. Here are 5 major upgrades to help you create, iterate and collaborate: 🎨 AI-Native Canvas 🧠 Smarter Design Agent 🎙️ Voice ⚡️ Instant Prototypes 📐 Design Systems and DESIGN.md Rolling out now. Details and product walkthrough video in 🧵

@MingtaKaivo @victorialslocum How does one addresses these challenges and what are the mitigations for these issues like syncs across?

@victorialslocum the observation most miss: shared memory is harder than shared tools. each agent caching its own context is cheap. syncing across N agents? that's where designs break.

Building a multi-agent system 𝗶𝘀𝗻'𝘁 𝗷𝘂𝘀𝘁 𝗮𝗱𝗱𝗶𝗻𝗴 𝗺𝗼𝗿𝗲 𝗮𝗴𝗲𝗻𝘁𝘀 (This is why specialized agents beat generalists every time) Instead of a single agent trying to handle everything, 𝗠𝘂𝗹𝘁𝗶-𝗮𝗴𝗲𝗻𝘁 𝘀𝘆𝘀𝘁𝗲𝗺𝘀 employ teams of specialized agents, each with its own focused task. So for example, you could have a team of: A 𝗣𝗹𝗮𝗻𝗻𝗶𝗻𝗴 𝗔𝗴𝗲𝗻𝘁 that decides how to handle the users request. A 𝗤𝘂𝗲𝗿𝘆 𝗥𝗲𝘄𝗿𝗶𝘁𝗶𝗻𝗴 𝗔𝗴𝗲𝗻𝘁 that takes messy user queries and decomposes them into more manageable, clear subqueries. A 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 𝗔𝗴𝗲𝗻𝘁 𝗮𝗻𝗱/𝗼𝗿 𝗗𝗮𝘁𝗮 𝗦𝗼𝘂𝗿𝗰𝗲 𝗦𝗲𝗹𝗲𝗰𝘁𝗼𝗿 that specializes in finding the right information from the right source. A 𝗧𝗼𝗼𝗹 𝗥𝗼𝘂𝘁𝗶𝗻𝗴 𝗔𝗴𝗲𝗻𝘁 that decides which tools to use and when. A 𝗔𝗻𝘀𝘄𝗲𝗿 𝗔𝗴𝗲𝗻𝘁 that decides how to best combine all the results to provide the more complete answer to the user. 𝗠𝗲𝗺𝗼𝗿𝘆 is what allows an agentic system like this to work. Short-term memory tracks the current conversation and recent actions. Long-term memory stores patterns, successful strategies, and domain knowledge. When agents share memory, they build on each other's work instead of starting from scratch every time. Each agent has access to specific tools. The retrieval agents can call different search APIs. The validation agent might use a scoring model. The synthesis agent has access to the LLM for generation. They don't all need every tool - they just need the right ones for their specialized task. IMHO, this is way more robust than a single agent trying to handle everything. When retrieval fails, the coordinator can try a different retrieval agent. When validation catches low-quality results, it can trigger a re-retrieval with different parameters. Specialization means better error handling and more reliable outcomes. More agents means more complexity. But for complex tasks, multi-agent systems consistently outperform single agents trying to do it all.

This combo always slaps!!🧋🍔 #mcdonalds

@andrewchen This is so true! But tbh I am having fun ngl

AI is supposed to save me time, but now I find myself building stuff all evening and weekend and it's actually increasing my time in front of the computer WTF

@nalinrajput23 Yes I can believe since there's no purple vibe-coded shade on the UI

can you believe they made this without claude

build agents with LangSmith and let them execute code securely. We’re launching LangSmith Sandboxes today!

I just published LangChain and NVIDIA Are Building the Infrastructure Layer for Enterprise AI Agents medium.com/p/langchain-an…

@thesayannayak 🌚🌚

Do you actually revisit your Bookmarks on 𝕏 ?👀

@adahstwt GoDaddy

be honest, where do you usually buy your domains? - GoDaddy - Cloudflare - Dynadot - Namecheap - Porkbun Drop your websites, let's have a look👇🏻

@marlowxbt I can smell more stories like these in coming days🤨🙄

AWS sent me a $47 bill. I haven't used AWS in 8 months. Logged in to shut it down. Found one EC2 instance running. Micro. $0.0058 per hour. Someone spun it up in February using my old credentials I forgot to rotate. I was about to terminate it. Then opened the logs. A bot. Running 24/7 since February. Connected to Binance WebSocket and a prediction platform API. Executing trades every 3 minutes. I followed the wallet address from the config file. 0x732F1. $339,140 profit. 38,945 predictions. Joined February 2026. Bio: there are no socials/websites related to this profile. → Wallet: t.me/PolyGunSniperB… Someone used my forgotten $47/month server to run a bot that made $339K. 38,945 trades. 800 per day. BTC moves on Binance. Platform lags 25 seconds. Bot buys old price. Collects $1. Repeat. The code was 26 lines of Python. Clean. No comments. No readme. Just a WebSocket listener, a price comparison and a buy function with a 15 second sleep timer. $339K profit on a $47 monthly server bill. ROI on the server alone: 721,574%. I checked the SSH login history. One IP address. Vietnam. Logged in once in February. Never again. Set the bot. Left. Someone halfway across the world found my exposed credentials, didn't steal my data, didn't mine anything. Just quietly parked a 26 line script on my cheapest server and let it print. I didn't terminate the instance. Changed the password. Sat there reading the logs for 2 hours. The bot is still running. The wallet is still active. $113K in open positions right now. My $47 AWS bill just became the most profitable invoice I never meant to pay.

@rishikagupta__ Observe, Evaluate, Design more reliability, Scalability, Pptimization...

If everything can be automated with AI, what will humans do?

@fidexcode "Just in case"

What is the point of making the interview harder than the actual job?

@rxhit05 Same goes for job seeking people in 2026 Without showcasing work in public, talent & opportunities die in silence

If you’re a founder: Marketing and selling matter more than building. Great products die in silence.

@praveenTweets @Uber "AI will replace Software Engineers" is only half right. It'll replace typing out code. Not the judgment of whether that code is correct, scalable, or even solving the right problem. Evaluation skills > coding skills in the near future.

Agentic software engineering adoption is on fire at @Uber. 1,800 code changes per week are now written entirely by Uber's internal background coding agent, and 95% of our engineers now use AI every month across all the tools we track. This is a real reset moment for engineering; it's one of the most exciting times to lead. This shift requires builders to be curious and hands-on. I’m incredibly lucky to be surrounded by a team that’s doing exactly that. The best part is that the strongest adoption isn’t being pushed top down from leadership announcements; it’s coming from engineers who are quietly experimenting, quietly shipping, and quietly pushing things forward. I love spending time with those engineers because there’s no substitute for being close to the work. Over the last few months, we leaned in hard, and the results have been phenomenal. The bigger shift: going agentic. 84% of AI users are now working with agent-style workflows, not just tab completion. Claude Code usage nearly doubled in 2 months (32% → 63%), while IDE-based tools have largely plateaued. Engineers are moving from accepting suggestions to delegating tasks. Even within traditional IDEs, ~70% of committed code is now AI-generated. Background agents are writing code autonomously. Our internal background coding agent went from <1% of all code changes to 8% in just a few months. There is zero human authoring. Engineers review and approve, but the code is written entirely by AI agents. The role of the engineer is shifting - from writing every line to architecting systems and reviewing AI-generated code. More to come from the @UberEng team in the coming days.

An AI-powered ad bidding platform that uses a LangGraph conversational agent to make real-time auction decisions, optimize budgets, and simulate strategies. Backed by a full production stack Kafka for live bid streaming, Snowflake for data, LangSmith for evaluation, and Kubernetes for deployment.

What are you building this week?

@Thepromoplay Ikr! Totally agreed! However, I've started reading more articles on X Few resonate with the work I do, few don't So I selectively keep the tech updates that I understand or are remotely close to what I do. Rest I filter out! That's my approach. Keeps me sane😅

@icedcoffeecoder Honestly? I don't try to keep up with everything — I would drown. The secret for me: accept that you will miss things. FOMO kills productivity.😅 What is your approach?

How do you keep up with tech updates everyday ?🤔

@fidexcode @Jesse_can_code npm ai

@Jesse_can_code When are you building your first AI model

Day 1 learning machine learning

@sachinyadav699 I like how they suggesting you email id for login🙃

$100k AI startup - fully vibe coded - register page leaking user emails - investors love this

@craigzLiszt this is like THE most random tweet I’ve seen

started learning hindi, just in case

Multimodal evals are going to unlock more advanced take on exisiting observability layer!