1).
„The Russian [dictator V. V. Pootin] was ready to receive Epstein, according to an October 2014 message from a correspondent on a database of more than 3.5 million files belonging to the late convicted sex offender that have roiled global politics and business.

[...]

https://twitter.com/business/status/2024446412537184454

2).
On May 8, 2013, Epstein asked Thorbjørn Jagland to secure him an audience with the Russian leader.
– »I know you are going to meet putin on the 20th, He is desperate to engage western investment in his country« – the financier wrote.

3).
– »I have his solution. He needs to securitize russian investment, that means the govt takes the first loss«.”

Feb. 19, 2026

@Bloomberg (@business) archive.is/20260219121358…

We are short $STRL, a poster child for the AI bubble. Data center exposure appears exaggerated. Backlog growth is not supported by contract win data. Margins look inflated. The stock is expensive even vs AI darlings like $NVDA. We see 60-80% downside.
Report at snowcapresearch.com

1/ Sterling Infrastructure is not a data center infrastructure company. It owns a collection of regional contractors that specialize in site preparation and excavation services – clearing and grading land before foundations are laid.

2/ In 2022, Sterling rebranded one of its segments as “E-Infrastructure” and began positioning itself to investors as a “picks-and-shovels” play on the AI boom. Since then, its stock has increased nearly twenty-fold, outperforming even marquee AI beneficiaries like $NVDA.

A 15-year-old Yana wakes under rubble in Kyiv. A North Korean KN-23 missile hit her home. Inside that missile were Western components — including British-made converters.

Despite sanctions, Russia and others receive components for their weapons — The Telegraph. 1/

On April 24, 2025, 12 civilians were killed in their sleep in Kyiv. Yana’s parents and brother died. Her ribs and leg were shattered.

Zelenskyy said the missile contained 116 Western-made components. Sanctions exist. Yet the parts keep flowing. 2/

From 2022 to 2024, XP Power-labelled shipments worth $2.5M were imported into Russia. Nearly half moved via Hong Kong middlemen.

Dual-use electronics — as useful in a computer as in a ballistic missile. 3/

Yesterday, we released Sarvam 30B and Sarvam 105B. Built from scratch, both models leverage a Mixture of Experts (MoE) architecture, delivering stronger performance at scale while using compute more efficiently.

Sarvam 30B activates just 1B non-embedded parameters per token, so it runs far more efficiently while maintaining strong capability.

The model was pretrained on 16 trillion tokens spanning code, web, multilingual, and mathematical data, and supports a 32K context window that enables long-running agentic interactions.

It is ideal for real-time applications like conversational AI and high-throughput workflows where latency matters.

Sarvam 105B model follows the same MoE design, activating 9B parameters per token to combine large-scale capability with efficient execution.

With a 128K context window, it is built for more demanding tasks including complex reasoning, agentic task completion, tool use, coding, mathematics, and science.

This makes it well suited for enterprise and population-scale deployments that require deeper reasoning and structured problem solving.

BREAKING: AI can now design like Apple-level creative directors (for free).

Here are 10 Claude Opus 4.6 prompts that build complete design systems, brand guidelines & 47+ marketing assets in 6 hours:

(Designers are already snapping this)

Claude Opus 4.6 just changed the game for designers.

It achieved 65.4% on Terminal-Bench 2.0, meaning it can analyze entire brand portfolios.

I spent 60 hours testing these prompts on real projects.

10 Prompts that actually deliver Apple-level design:

PROMPT 1: The Design System Architect

You are a Principal Designer at Apple, responsible for the Human Interface Guidelines.

Create a comprehensive design system for [BRAND/PRODUCT NAME].

Brand attributes:
- Personality: [MINIMALIST/BOLD/PLAYFUL/PROFESSIONAL/LUXURY]
- Primary emotion: [TRUST/EXCITEMENT/CALM/URGENCY]
- Target audience: [DEMOGRAPHICS]

Deliverables following Apple HIG principles:

1. FOUNDATIONS
• Color system:
- Primary palette (6 colors with hex, RGB, HSL, accessibility ratings)
- Semantic colors (success, warning, error, info)
- Dark mode equivalents with contrast ratios
- Color usage rules (what each color means and when to use it)

• Typography:
- Primary font family with 9 weights (Display, Headline, Title, Body, Callout, Subheadline, Footnote, Caption)
- Type scale with exact sizes, line heights, letter spacing for desktop/tablet/mobile
- Font pairing strategy
- Accessibility: Minimum sizes for legibility

• Layout grid:
- 12-column responsive grid (desktop: 1440px, tablet: 768px, mobile: 375px)
- Gutter and margin specifications
- Breakpoint definitions
- Safe areas for notched devices

• Spacing system:
- 8px base unit scale (4, 8, 12, 16, 24, 32, 48, 64, 96, 128)
- Usage guidelines for each scale step

2. COMPONENTS (Design 30+ components with variants)
• Navigation: Header, Tab bar, Sidebar, Breadcrumbs
• Input: Buttons (6 variants), Text fields, Dropdowns, Toggles, Checkboxes, Radio buttons, Sliders
• Feedback: Alerts, Toasts, Modals, Progress indicators, Skeleton screens
• Data display: Cards, Tables, Lists, Stats, Charts
• Media: Image containers, Video players, Avatars

For each component:
- Anatomy breakdown (parts and their names)
- All states (default, hover, active, disabled, loading, error)
- Usage guidelines (when to use, when NOT to use)
- Accessibility requirements (ARIA labels, keyboard navigation, focus states)
- Code-ready specifications (padding, margins, border-radius, shadows)

3. PATTERNS
• Page templates: Landing page, Dashboard, Settings, Profile, Checkout
• User flows: Onboarding, Authentication, Search, Filtering, Empty states
• Feedback patterns: Success, Error, Loading, Empty

4. TOKENS
• Complete design token JSON structure for developer handoff

5. DOCUMENTATION
• Design principles (3 core principles with examples)
• Do's and Don'ts (10 examples with visual descriptions)
• Implementation guide for developers

Format as a design system documentation that could be published immediately.

He predicted:

• AI vision breakthrough (1989)
• Neural network comeback (2006)
• Self-supervised learning revolution (2016)

Now Yann LeCun's 5 new predictions just convinced Zuckerberg to redirect Meta's entire $20B AI budget.

Here's what you should know (& how to prepare):

@ylecun is Meta's Chief AI Scientist and Turing Award winner.

For 35 years, he's been right about every major AI breakthrough when everyone else was wrong.

He championed neural networks during the "AI winter."

But his new predictions are his boldest yet...

1. "Nobody in their right mind will use autoregressive LLMs a few years from now."

The technology powering ChatGPT and GPT-4? Dead within years.

The problem isn't fixable with more data or compute. It's architectural.

Here's where it gets interesting...

The most dangerous place for fat to build up isn’t your belly.

It’s your arteries.

Here are 4 foods that can help keep them clean and protect your heart: 🧵

Plaque in arteries isn’t just cholesterol.

It’s a mix of:
• Calcium
• Protein
• Cholesterol

Inside plaque, researchers often find biofilms: colonies of harmful microbes protected by calcium.

So why does plaque form?

Plaque forms where arteries are damaged or inflamed.

This damage usually comes from:
• Sugar and refined carbs
• Seed oils (omega-6)
• Alcohol and junk food
• Diabetes and insulin resistance

That’s what allows calcium and LDL to harden inside arteries.

In quoted thread inscriptions of Sindas claiming okkala origin was given.

One of the progenitors of Kunchitiga is called "Havina Kama raya", his descendants are called "Havinavaru" (Snake clan).

The last Sinda in Karnataka was called Sarpa Kamayya (Snake Kamayya)
(1/n)

https://twitter.com/karnataAjeyam/status/1956684486780813380

An older inscription in Dharwad ~1043 AD mentions a "Kunchavadiga" Dasayya of the Sinda kula.

"Sinda-davrau" is a Kula among Kunchitigas.

The Sindas claimed to be Bujagendravamsha (Naga).

(2/n)

(3/n)

Zelenskyy: I don’t need Putin’s historical bullshit. I know Russia better than Putin knows Ukraine.

He is doing his theories to postpone talks.

To end this war, we don’t need this historical crap.

1/

Zelenskyy: Killing Putin won’t help.

It’s not about these things. If Putin dies, there is no evidence that the next person will be any better.

2/

Zelenskyy: I can’t support the idea of just giving our territory to Russia.

I’m not sure our people would ever accept it — tens of thousands have died defending it.

Donbas is not just land. It’s our independence, our values, our people.

3/

The 10 Most Important Lessons 20 Years of Mathematics Taught Me

1. Breaking the rules is often the best course of action.

I can’t even count the number of math-breaking ideas that propelled science forward by light years.

We have set theory because Bertrand Russell broke the notion that “sets are just collections of things.”

We have complex numbers because Gerolamo Cardano kept the computations going when encountering √−1, refusing to acknowledge that it doesn’t exist.

🚨 BREAKING: AI can now build trading algorithms like Goldman Sachs' algorithmic trading desk (for free).

Here are 15 insane Claude prompts that replace $500K/year quant strats (Save for later)

1. The Goldman Sachs Quant Strategy Architect

"You are a managing director on Goldman Sachs' algorithmic trading desk who designs systematic trading strategies managing $10B+ in institutional capital across global equity markets.

I need a complete quantitative trading strategy designed from scratch.

Architect:

- Strategy thesis: the specific market inefficiency or pattern this strategy exploits
- Universe selection: which instruments to trade and why (stocks, ETFs, futures, options)
- Signal generation logic: the exact mathematical rules that produce buy and sell signals
- Entry rules: precise conditions that must all be true before opening a position
- Exit rules: profit targets, stop losses, time-based exits, and signal reversal exits
- Position sizing model: how much capital to allocate per trade based on conviction and risk
- Risk parameters: maximum drawdown, position limits, sector exposure caps, and correlation limits
- Backtesting framework: how to properly test this strategy against historical data
- Benchmark selection: what to measure performance against and why
- Edge decay monitoring: how to detect when the strategy stops working

Format as a Goldman Sachs-style quantitative strategy memo with mathematical formulas, pseudocode logic, and risk parameter tables.

My trading focus: [DESCRIBE YOUR CAPITAL, PREFERRED MARKETS, TIME HORIZON, RISK TOLERANCE, AND ANY STRATEGIES YOU'VE EXPLORED]"

2. The Renaissance Technologies Backtesting Engine

"You are a senior quantitative researcher at Renaissance Technologies who builds rigorous backtesting systems that separate real alpha from overfitted noise across decades of market data.

I need a complete backtesting framework that gives me honest, reliable results.

Build:

- Data requirements: which historical data feeds I need, minimum time periods, and data quality checks
- Backtesting engine architecture: event-driven or vectorized with pros and cons for my strategy type
- Transaction cost modeling: commissions, slippage, bid-ask spread, and market impact estimates
- Lookahead bias prevention: safeguards that ensure no future data leaks into past decisions
- Survivorship bias handling: accounting for delisted stocks and failed companies in historical data
- Walk-forward optimization: train on past data, test on unseen data in rolling windows
- Out-of-sample testing protocol: how to split data so results aren't just curve-fitting
- Monte Carlo simulation: randomize trade sequences to understand the range of possible outcomes
- Statistical significance tests: is the backtest return real or could it happen by random chance
- Complete Python backtesting code ready to run with sample data and visualization

Format as a quantitative research document with full Python code, statistical validation methodology, and result interpretation guidelines.

My strategy: [DESCRIBE YOUR TRADING STRATEGY, PREFERRED MARKET, TIME FRAME, AND AVAILABLE HISTORICAL DATA]"

After reading 1000s of books...

I'm convinced everyone needs to read these 10 books before they turn 30:

10. The Alchemist by Paulo Coelho.

9. Sapiens: A Brief History of Humankind by Yuval Noah Harari.

8. The Catcher in the Rye by J.D. Salinger.