Tech With Gul

  • Home
  • Tech With Gul

Tech With Gul Tech enthusiast sharing AI, tools, and trends. Insights, updates, and tech news.

02/06/2026

Google just open-sourced an AI model that can predict the future. 🔮

It’s called TimesFM — a Time Series Foundation Model trained on 100 billion real-world data points.

📈 Sales forecasts
🚗 Traffic predictions
⚡ Energy demand
🏥 Health metrics
💰 Business planning

The crazy part?

You don’t need to train a custom model anymore.

Just provide historical data and TimesFM generates forecasts in minutes.

The latest version doesn’t just give one prediction — it gives:
✅ Best-case scenario
✅ Worst-case scenario
✅ Most likely outcome

And it’s already integrated with Google Sheets and BigQuery.

This is a massive step toward making forecasting accessible to everyone.

Comment **PREDICT** and I’ll send you the direct link to try it.

Follow for daily AI updates, tools, tutorials, and practical AI use cases.

28/05/2026

93 AI agents built a full OS in 12 hours. For under $1,000. 🤯

Google ne Google I/O 2026 mein jo reveal kiya — woh sab expecting nahi kar rahe the.

Gemini 3.5 Flash — Google ka naya model — aur yeh already unke apne Pro model ko beat kar raha hai.

Real-world tasks par. Coding par. Agent benchmarks par.

Cheaper model ne expensive model ko dethrone kar diya.

Aur phir unhone live demo diya:
→ 93 parallel sub-agents
→ 15,000+ model calls
→ Ek poora functioning OS — 12 hours mein

Antigravity 2.0 ke saath — aap sirf goal define karte ho.
Flash baaki sab handle karta hai. Simultaneously. ⚡

Comment FLASH and I’ll send you the full breakdown link 👇



emini 3.5 Flash, Gemini Flash vs Pro, Google AI agents, Antigravity 2.0, multi agent AI, Gemini 3.5 Flash benchmark, Google agentic AI, AI builds OS, 93 AI agents, Gemini Flash speed, Google AI 2026, Google DeepMind Gemini, agentic AI platform, AI coding agents, Gemini API, Android Studio Gemini, Google AI tools, tech news Pakistan, techwithgul, AI agents 2026

27/05/2026

Eid Mubarak everyone ✨

This Eid, don’t let your amazing memories stay dull or low quality.
I used ONE AI prompt to transform my normal photo into a cinematic DSLR-style picture with better lighting, colors, depth, and professional vibes 📸

Perfect for:
✨ Eid portraits
✨ Family pictures
✨ Instagram reels
✨ DSLR-style edits
✨ AI photo enhancement
✨ Cinematic videos & photos

Comment “PROMPT” below and I’ll send you the exact AI prompt 👇

AI photo editing, DSLR photo effect, cinematic photo editing, Eid photography ideas, mobile photography, AI prompt for photos, professional photo enhancement, Instagram reel editing, viral AI editing, cinematic lighting

23/05/2026

AI video just changed forever.

Google dropped Gemini Omni at I/O 2026, and it fixes the one thing every AI video tool was getting wrong.

Text. Math. Writing on screens.

Runway, Sora, Kling — they all blurred letters, scrambled equations, broke words into random pixels.

Gemini Omni doesn’t just render text. It *understands* what it’s drawing.

Professor solving trigonometry on a chalkboard → math was correct. Every symbol. Every step.

And the wildest part?

Video editing is now just conversation.

“Replace the background.”
“Make the lighting cinematic.”
“Remove that object.”

Done. No timeline. No keyframes. No headaches.

This is intelligence inside video, not just generation.

Comment **OMNI** below and I’ll send you the full breakdown 👇

—



Google Gemini, AI video generation, text to video, Claude, artificial intelligence, video creator tools, future of AI

21/05/2026

Mouse pointer ab sirf click nahi karega…

Yeh aapki baat samjhega. 🤯

Google DeepMind ne AI ko seedha mouse mein daal diya hai.

Ab sirf point karo aur bolo — „yeh meeting 8 PM kar do“ — aur computer khud kar dega.

Koi shortcuts yaad karne ki zaroorat nahi.
Koi menu dhundne ki zaroorat nahi.
Sirf bol do.

Yeh sirf ek feature nahi hai…
Yeh pehla step hai jahan computer commands nahi — human intention samjhega. 👀

Aapko kya lagta hai — kya aap aise AI ko trust karoge jo aapki poori screen dekh sakta ho?
Comment mein batao 👇



Google Magic Pointer, AI mouse pointer, Google DeepMind Magic Pointer, Google AI pointer, Gemini AI pointer, Googlebook laptop, Google AI 2026, Google I/O 2026, AI computer features, DeepMind AI mouse, Google Gemini features, AI operating system, future of computing, Google AI update, tech news Pakistan, AI Pakistan, Google AI Urdu, techwithgul, AI tools 2026, Google AI laptop

19/05/2026

yPal, flags every mismatch
↳ Restaurant: 30-day cash flow forecast + supplier payment alerts
↳ Retail: clearance event? Claude pulls your data, writes the strategy, generates Canva assets

And if you’re in Pakistan 🇵🇰 — this is for you.

SMEs make up 90% of our economy. We’ve been running on WhatsApp, spreadsheets, and gut feel while larger companies automated everything. These tools work from anywhere — including here.

Comment SMALL and I’ll send you the full setup guide — including how to use it from Pakistan. 👇

claude, ai, anthropic, claude ai, ai tools, small business, business automation, ai for business, quickbooks, canva, entrepreneur, pakistan business, sme pakistan, ai 2025, business tools, cash flow, ecommerce, ai workflow, paypal, hubspot

15/05/2026

Your AI is wasting 71× more tokens than it needs to on every single query.

Not because the model is bad.

Because it’s reading your entire codebase from scratch — every time — with no idea which files matter, what connects to what, or how your architecture fits together.

Graphify fixes this.

One run. It parses your code, docs, PDFs, images, and videos — and turns everything into a traversable knowledge graph.

Your AI stops searching keywords. It navigates structure.

Result: 71.5× fewer tokens per query. 30,000 GitHub stars in 10 days. Works natively with Claude Code, Cursor, Gemini CLI, and Codex.

One install command. Commit the graph. Your whole team starts with the map already built.

Comment GRAPH and I’ll send you the full setup guide.

ClaudeCode, AITools, AIAgent, CursorAI, OpenSource, GitHub, KnowledgeGraph, TokenOptimization, LLM, PromptEngineering, DeveloperTools, BuildWithAI, CodeSmarter, CodingTips, DevLife, SoftwareEngineering, AIForDevelopers, TechTools, AIProductivity, TechWithGul

13/05/2026

Anthropic just gave every finance team a 10-person AI department. Pitch decks, KYC files, month-end close — there’s a named agent for each one.

Here’s what shipped:

Research side: → Pitch Builder — target lists, comps, pitchbook drafts → Meeting Preparer — client and counterparty briefs before calls → Earnings Reviewer — reads filings, updates models, flags thesis changes → Model Builder — financial models from filings and data feeds → Market Researcher — sector tracking and broker research synthesis

Operations side: → Valuation Reviewer — checks valuations against comps and methodology → GL Reconciler — reconciles ledger accounts, runs NAV calculations → Month-End Closer — runs the close checklist, prepares journal entries → Statement Auditor — reviews financials for audit-readiness → KYC Screener — assembles entity files, packages compliance escalations

All of them work inside Excel, PowerPoint, Word, and Outlook — and context carries between apps. Build a model in Excel, open PowerPoint, Claude already knows the numbers.

Deploy as a plugin alongside the analyst — or as a Managed Agent running autonomously overnight, with a full audit log for compliance.

64.37% on Vals AI’s Finance Agent benchmark. Industry lead.

Comment PITCH and I’ll send you the full breakdown 👇

anthropic finance agents, claude finance, ai finance agents, pitch builder ai, kyc screener ai, month-end close automation, claude excel, claude powerpoint, microsoft 365 ai, managed agents, claude cowork, claude code finance, fintech ai, valuation ai, financial model ai, earnings reviewer, gl reconciler, statement auditor, market researcher ai, claude dispatch, techwithgul, anthropic 2026, vals ai benchmark, moody’s mcp, finance automation, ai for bankers, llm finance, context engineering finance, agent templates, connectors, subagents

12/05/2026

Prompt engineering was wave one.

Wave two already created four new engineering fields — and if you want to build a real career in AI, these are the ones you need to know.

① Context Engineering What goes into an agent’s context window — and what gets cut. Most AI failures aren’t model failures. They’re context failures. This field fixes that.

② Memory Engineering What the agent saves between sessions. Short-term vs. long-term storage, and a retrieval layer that surfaces the right thing at the right time. Done right, it gets smarter every run — not starting from zero.

③ Harness Engineering Agent = Model + Harness. The harness controls when to call tools, how to recover from errors, how to break out of loops. Building an agent is easy. Making it reliable in production is the actual job.

④ Evaluation Engineering Building the test suites that prove your agent works before it ships. LangChain’s 2026 report: quality is the #1 barrier to deployment — cited by 32% of engineering teams. Most teams are still shipping without proper evals. That’s exactly why this field exists.

Four fields. All hiring right now.

Comment CONTEXT and I’ll send you the full breakdown + resources for each field 👇

context engineering, memory engineering, harness engineering, evaluation engineering, ai engineering, ai agents, ai jobs, ai careers, prompt engineering, llm engineering, ai agent development, ai for developers, future of work, ai engineering fields, agent harness, context window, agent memory, agent evals, ai evaluation, rag, retrieval augmented generation, ai loop, techwithgul, ai productivity, wave two ai, anthropic, langchain, martin fowler, ai systems, ai observability

06/05/2026

GitHub just ranked this #1 — and almost nobody has it set up yet.

It’s called Ruflo. You spin up 60+ AI agents that all work together inside Claude Code. They share memory. They learn from every job. And they’re completely free.

Here’s what makes it different:

→ Queen agents manage the team
→ Tactical agents handle coding, testing, research, and review
→ Shared memory across every session — they remember everything
→ SONA self-learning — every run makes the next one faster
→ Routes simple tasks to cheap models, complex ones to powerful models

Result: 84.8% on SWEBench. 75% fewer API calls. 250% more Claude usage from the same plan.

Setup is one line:
`claude mcp add ruflow`

That’s it. You now have an AI workforce.

45,000 GitHub stars. 100,000 users a month. 80+ countries. Completely free.

Comment **FLOW** and I’ll send you the full setup guide 👇

ruflo, ruflow, claude code, ai agents, multi agent ai, agent swarms, ai tools, open source ai, github trending, free ai tools, ai automation, ai workforce, developer tools, coding tools, anthropic, claude ai, ai productivity, swarm intelligence, ai coding, techwithgul, ai for business, no code ai, ai setup, ruven cohen, agent orchestration, SWEBench, ai memory, federation ai, local ai, ollama, ai workflow

Address


Website

Alerts

Be the first to know and let us send you an email when Tech With Gul posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Shortcuts

  • Want your business to be the top-listed Engineering Company?

Share