Trading Socrates

muslim ~ poet ~ philosopher ~ strategist ~ optimizer ~ synergist ~
NOSTR: npub1vhfl0dms4gpspr77hcd3unm0ycf06phfpr7p9lwtsynwzqe0d8nqafulkj

[Trading Socrates @dravitch]

https://t.co/FiyPtSYu4O

[Trading Socrates @dravitch]

I stopped using Claude Code (paid subscription). And I went all in with OpenCode as client agentic and Nvidia Nemotron 3Super Free as LLM. The combinaison est crazy for coding. Full dev complète in 48h, no question, no mistake, ni rate limit, deep focus. OpenCode is 10$/m.

[Erick @ErickSky]

¿Sabes cuál es la casa más loca que se está construyendo ahora mismo en Estados Unidos?

No es la más grande.
No es la más cara.

Es la que viene con su propio data center de IA pegado a la pared.

Nvidia, la empresa que domina la inteligencia artificial, se alió con Span (la startup de paneles eléctricos inteligentes) y PulteGroup (uno de los mayores constructores de casas de EE.UU.) para hacer algo que parece sacado de Black Mirror… pero es 100% real.

Se llaman unidades XFRA.

Cada una es una bestia: 16 GPUs Blackwell RTX PRO 6000 (las más potentes del planeta), enfriadas por líquido para que sean completamente silenciosas, 4 procesadores AMD EPYC y 3 terabytes de RAM.

Todo metido en una caja blanca del tamaño de un equipo de aire acondicionado que se instala al lado de tu casa.

Lo más brillante: Usan la electricidad que tu casa no está consumiendo gracias al panel inteligente de Span.

No hay que ampliar subestaciones ni pedir permisos eternos.

Se despliega 6 veces más rápido y sale 5 veces más barato que construir un data center gigante tradicional.

Y aquí viene la parte que te va a volar la cabeza:

- Te pagan una tarifa fija todos los meses por prestar tu luz y tu WiFi.

- Tu casa ya no solo te da techo y recuerdos. Ahora te da ingresos pasivos mientras procesa inteligencia artificial para el mundo entero.

- Mientras tú duermes, ves Netflix o cocinas, tu casa está trabajando como un nodo de la red global de IA.

Esto no es el futuro.

Esto ya está pasando en comunidades reales de PulteGroup.

Bienvenidos al 2026, donde hasta tu casa puede ser parte de la revolución de la inteligencia artificial… y encima te paga por ello.

[xIA @xiathis]

ANTHROPIC ACABA DE LANZAR UNA ESTRUCTURA DE BOT DE TRADING PARA MERCADOS DE PREDICCIÓN 🤯

TASA DE ÉXITO DEL 68.4%.

$300–$1,500 POR DÍA.

[gus @igus_ai]

Cancelé $2.000/mes en suscripciones de Trading

Reemplacé casi todo por repositorios Open-Source 100% gratis

Este es el stack completo:

1. TradingView Pro ($30/mes) → lightweight-charts

14K estrellas. Creado por el propio equipo de TradingView. 45KB. Gratis.
> https://t.co/VqpSa8RNuR

2. Bloomberg Terminal ($2.000/mes) → fredapi + Claude

Acceso a todos los datasets macroeconómicos publicados por la Fed mediante API gratuita
> https://t.co/1dvvJRkXVB

3. Plataforma de backtesting ($100/mes) → prediction-market-backtesting

Fork de NautilusTrader con adaptadores para Polymarket y Kalshi
> https://t.co/wzFhoGQNbG

4. Ingeniería inversa de estrategias → polybot

Infraestructura de ejecución y datos de mercado con paper trading.
Kafka, ClickHouse y Grafana como pipeline completo de analíticas
> https://t.co/x3rufeBuyX

5. Paper trading para agentes IA → polymarket-paper-trader

Order books reales, modelo exacto de fees y tracking de slippage tu agente de Claude recibe $10K ficticios para operar
> https://t.co/kp9IZyacpF

6. Ahorro de tokens → rtk

Proxy CLI que reduce entre un 60-90% el consumo de tokens en Claude Code
escrito en Rust, binario único y compatible con 10 herramientas IA
> https://t.co/9n4E6OdxA6

7. Claude Code ($200/mes) → goose

35K estrellas. Desarrollado por Block (Jack Dorsey). Escrito en Rust. Funciona con cualquier LLM y ofrece un loop completo de agentes IA
> https://t.co/S8SDZjNbwz

Antes: +$2.600/mes
Ahora: prácticamente $0

Guárdate este post, me lo agradecerás. 🔖

[Farza 🇵🇰🇺🇸 @FarzaTV]

@karpathy Also, I'll here's the skill I made for the wiki.

If you wanna try yourself + hack around with it load it up in your agent of choice. Should work for nearly any data source (Notion, iMessage, etc)

https://t.co/Cj9Bi8yPi7

[Trading Socrates @dravitch]

@cyrilXBT Behind your Claude Code optimisation, there is a knowledge management framework that links all agents and all LLM. You just need to formalize it properly to make millions.

[wincy.eth @gusik4ever]

the fastest growing GitHub repos in finance this week:

1. TauricResearch/TradingAgents (+2.5K ★)

simulates a full trading firm with LLM agents. one researches, one manages risk, one makes the call and they argue before every trade.

2. disler/last30days-skill (+2K ★)

AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket and the web. drop it into any Claude-compatible setup and get instant deep-dive research on anything.

3. TauricResearch/TradingAgents-CN (+1K ★)

Chinese-enhanced fork of TradingAgents. same multi-agent LLM trading architecture, fully localized for Chinese markets and data sources. 23K stars and climbing.

4. OpenBB (+1K ★)

financial data platform for analysts, quants and AI agents. the open-source Bloomberg alternative that keeps getting better every week.

5. furutech/daily_stock_analysis (+924 ★)

LLM-powered stock analyzer for US, A-share and H-share markets. real-time news + multi-source data + decision dashboard. runs on a schedule at zero cost. pure automation.

6. microsoft/qlib (+638 ★)

AI-oriented quant investment platform from Microsoft. covers the full pipeline from data to live trading. deep learning, auto-quant, backtesting — all in one place.

7. anthropics/claude-scientific-skills (+573 ★)

ready-to-use agent skills for research, science, engineering, finance and writing. plug-and-play toolkit for anyone building on top of Claude.

8. valuecell/valuecell (+315 ★)

community-driven, multi-agent platform for financial apps. still early but the architecture is solid and the use cases are stacking up fast.

9. e2b-dev/500-AI-Agents-Projects (+256 ★)

curated collection of 500 AI agent use cases across industries. the best reference list if you're figuring out what to build next.

10. Jon-Becker/prediction-market-analysis (+246 ★)

framework for collecting and analyzing prediction market data. includes the largest public dataset of Polymarket + Kalshi trades. researchers are already publishing papers on top of it.

bookmark this and start today.

[Trading Socrates @dravitch]

@marryevan999 No verifiable blueprint, no tests done, no code on github. Bangladeshi Marry is suspicions...

[Trading Socrates @dravitch]

@ImadDebrous_ Quel gros problème prioritaire résolu! Enfin plus d'embouteillage, plus de vie chère, plus prière de rue. C'était la condition du développement du Québec. Vive le Québec libre (des islamistes nudistes) !

[Trading Socrates @dravitch]

Iran a-t-il le droit de se défendre ?

[CG @cgtwts]

pov: Life after you start coding with Claude code

[Paweł Huryn @PawelHuryn]

I thought people were exaggerating. Then I burned $100+ of extra Claude usage in 1 hour.

What happened: agentic coding with Claude in Chrome enabled. 10-20x multiplier on already expensive operations.

A survival guide until the fix ships:

Claude Code (terminal):
→ npx @anthropic-ai/claude-code (bypasses the Bun cache bug)
→ Avoid --resume (breaks cache)
→ /compact to compress context mid-session
→ /effort medium for routine work, /effort low for simple tasks
→ --model sonnet for exploration. Opus for final passes only
→ /mcp — disable servers you're not actively using

VS Code / Cursor:
→ Same engine. Same bugs. Same fixes.

Cowork (desktop):
→ Default to Sonnet. Not every task needs Opus.
→ Auto-selects effort
→ Start a fresh session if context bloats

Dispatch (mobile):
→ No model picker in UI
→ Delete session, start new one if it spirals
→ You can ask it to dispatch sessions with specific models

The biggest lever most people miss: stop using Claude in Chrome for scraping. Dev-browser gives you DOM snapshots instead of screenshots. Text is 10-100x cheaper than pixels.

Anthropic knows. Fix is coming. Until then — /compact, /effort medium, and kill your unused MCPs.

[Trading Socrates @dravitch]

@claudeai What about the cost optimisation use ?

[Trading Socrates @dravitch]

@s_mohinii Bullshit. Claude Code is not a Local LLM.

[George Pu @TheGeorgePu]

Mistral just open-sourced a text-to-speech model that beats ElevenLabs.

3 GB of RAM. Runs locally. Free.

The thing people were paying per-word for last year runs on your laptop now.

[Nainsi Dwivedi @NainsiDwiv50980]

Most developers treat Claude Code like a smarter autocomplete.

That's the wrong mental model.

It's actually a 4-layer engineering system:

1️⃣ CLAUDE.md → persistent project memory
Architecture, rules, team conventions

2️⃣ Skills → auto-invoked knowledge packs
Testing patterns, code review, deploy workflows

3️⃣ Hooks → deterministic guardrails
Security checks, formatting, automation

4️⃣ Agents → specialized sub-agents
Break complex tasks into parallel workflows

Once these are configured properly:

Claude stops behaving like a chatbot.
It starts behaving like a senior engineer on your team.

Most people never reach this level because they skip the setup.

The gap between average AI output and production-level results isn't the model.

It's the infrastructure around it.

Here’s the full breakdown on exactly how to build this 👇

[Trading Socrates @dravitch]

Donnees: https://t.co/YVaObDC7om

[Gregor Zunic @gregpr07]

The infra god has spoken