惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

F
Fortinet All Blogs
S
Secure Thoughts
月光博客
月光博客
美团技术团队
雷峰网
雷峰网
Exploit-DB.com RSS Feed
Exploit-DB.com RSS Feed
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
N
News and Events Feed by Topic
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
Forbes - Security
Forbes - Security
W
WeLiveSecurity
P
Proofpoint News Feed
阮一峰的网络日志
阮一峰的网络日志
爱范儿
爱范儿
G
GRAHAM CLULEY
cs.AI updates on arXiv.org
cs.AI updates on arXiv.org
AI
AI
Last Week in AI
Last Week in AI
Google Online Security Blog
Google Online Security Blog
Schneier on Security
Schneier on Security
云风的 BLOG
云风的 BLOG
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Recent Announcements
Recent Announcements
Webroot Blog
Webroot Blog
T
Tor Project blog
Cisco Talos Blog
Cisco Talos Blog
N
News and Events Feed by Topic
罗磊的独立博客
The Register - Security
The Register - Security
Blog — PlanetScale
Blog — PlanetScale
T
Threat Research - Cisco Blogs
博客园 - 【当耐特】
Apple Machine Learning Research
Apple Machine Learning Research
人人都是产品经理
人人都是产品经理
T
The Exploit Database - CXSecurity.com
www.infosecurity-magazine.com
www.infosecurity-magazine.com
B
Blog
腾讯CDC
Microsoft Azure Blog
Microsoft Azure Blog
酷 壳 – CoolShell
酷 壳 – CoolShell
H
Hacker News: Front Page
Application and Cybersecurity Blog
Application and Cybersecurity Blog
Engineering at Meta
Engineering at Meta
Latest news
Latest news
IT之家
IT之家
D
DataBreaches.Net
博客园 - 司徒正美
N
Netflix TechBlog - Medium
V
V2EX
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant Common SOC 2 Failures (Real World) Stop Vibe-Checking Your AI App: A Practical Guide to Evals How to Use SonarQube and SonarScanner Locally to Level Up Your Code Quality Your Next To-Do App Is Dead — I Replaced Mine with an OpenClaw AI Sign a Nostr event in 60 lines of Python using coincurve — no nostr-sdk, no nbxplorer, no rust toolchain ITGC Audit Explained Like You’re in Big 4 Patch Tuesday abril 2026: Microsoft parcha 163 vulnerabilidades y un zero-day en SharePoint Stop scraping everything: a better way to track competitor price changes Listing on MCPize + the Official MCP Registry while routing payments OUTSIDE the marketplace — how I kept 100% of my x402 revenue Building an AI-Powered Risk Intelligence System Using Serverless Architecture Why We Ripped Function Overloading Out of Our AI Toolchain Testing AI-Generated Code: How to Actually Know If It Works SaaS Churn Is Killing Your Business. Here Is What to Do About It (Without a Support Team) The Speed of AI Is No Longer Linear - And Self-Improving Models Are Why How to Implement RBAC for MCP Tools: A Practical Guide for Engineering Teams From Standard Quote to Persuasive Proposal: AI Automation for Arborists I built a CLI that scaffolds complete multi-tenant SaaS apps Axios CVE-2025–62718: The Silent SSRF Bug That Could Be Hiding in Your Node.js App Right Now The dashboard that ended our friendship Data Pipelines Explained Simply (and How to Build Them with Python) The Hidden Cost of AI Systems Nobody Talks About. undefined vs undeclared, and how typeof behaves Switching from file-based jobs to NATS/Kafka in Rust without changing code io_uring Adventures: Rust Servers That Love Syscalls Why Agentic AI is Killing the Traditional Database The POUR principles of web accessibility for developers and designers Quantum Neural Network 3D — A Deep Dive into Interactive WebGL Visualization How To Install Caveman In Codex On macOS And Windows Automation Pipeline Reliability: Why Your Workflow Breaks When Nobody Is Watching I Built an 'Open World' AI Coding Agent — It Works From ANY Folder From Freelancing to Product: A Tech Service Company's SaaS Transformation China's AI Giants: Adding Tencent Hunyuan & ByteDance Doubao to AI University (74 Providers) On the Vibe Coders and Their Lies clerk: Auto-Summarize Your Claude Code Sessions AI Weekly — 2026/04/10–04/17 | The Model Lockdown Is Here, but the Toolchain Is the Real Battleground AI 週報 — 2026/04/10–2026/04/17 模型封鎖潮來了,但工具鏈才是真戰場 Maybe this is how Open-Source apps are born... 🚀 Fine-Tune LLMs with LoRA and QLoRA: 2026 Guide tRPC v11 + Next.js App Router: End-to-End Type Safety Without the Boilerplate ShadCN UI in 2026: Why I Stopped Installing Component Libraries and Started Owning My Components SaaS Billing in React Server Components: Stripe + Supabase Without a Single `useEffect` Join our DEV Weekend Challenge — $1,000 in Prizes Across TEN winners! Submissions Due April 20 at 6:59 AM UTC. Implementing FSRS Spaced Repetition in Flutter + Supabase — Adding Memory Science to an AI Learning App "I Texted My Localhost From the Train — Claude Code Fixed the Bug Before I Got Home" I Built a Sales Prep AI and It Went Deeper Than Expected Design to Code #2: One JSON, Eleven Outputs Solving the 100M-Row Problem: A Summary Table Pattern for High-Volume Push Notification Logs Flutter Web With Wasm: What Actually Changes For Developers I Built 50 Royalty-Free Soundtracks for My Side Project in a Weekend Using AI Music Generation The Vibe Coding Security Checklist: 7 Things to Check Before You Ship Stop Letting Googlebot Guess Fix Your React App's SEO Right Desconstruindo o Streaming do LinkedIn: Como Criar um Engine de Extração de Vídeo de Alta Performance com HLS e FFmpeg (EDA Part-1) EDA (Exploratory Data Analysis) Explained With Real Life — Why Looking at Your Data Is the Most Important Step in Machine Learning Brand Relationship Management at Scale: Our 4-Touch Outreach System for 200+ Brands Why String.fromEnvironment() Might Return an Empty String in Dart JGuardrails 1.0.0 — Hardening Java LLM Apps Against Jailbreaks, Toxicity, and Prompt Injection Plan and Schedule a Full Week of Threads Content From One Claude Conversation Coding Cat Oran Ep3, Five Tables Changed Everything Updated: BFF Pattern I'm done watching freelancers get buried by 200 proposals. So I'm building the alternative. This is my first post BFS Algorithm in Java Step by Step Tutorial with Examples Tracking LLM Pricing Monthly: An Open Dataset for 22 AI Models How We Measure Content ROI on a Comparison Site: Revenue Attribution Without Perfect Data Introducing Nova AI Ops: The AI-Native Operating System for SRE Teams I built a free desktop video downloader for Windows — Grabbit How Talkie OCR Helps Vision-Impaired & Dyslexic Users Read the World Around Them VRCFaceTracking安装和iPhone面捕配置教程,有bug Even CrowdStrike Can't See Your Agents The Automation Gold Rush: What n8n Workflows and Claude Are Opening Up for Developers Right Now
# Criei um assistente que me manda os papers do dia traduzidos no Telegram
Lincoln Romais · 2026-05-28 · via DEV Community

Lincoln Romais

Se você acompanha pesquisa em IA, sabe que o HuggingFace Papers solta novos papers todo dia. O problema? São dezenas de abstracts em inglês, e ler tudo manualmente é inviável no dia a dia.

Resolvi automatizar isso: um script Python que busca os papers do dia, joga cada abstract pro meu LLM local via Ollama, traduz pra português e manda tudo pro Telegram. Sem pagar por API, sem nada na nuvem.

Neste artigo vou te mostrar como funciona e como você pode replicar em menos de 10 minutos.


O que o projeto faz

  1. Consulta a API do HuggingFace e pega os papers em alta do dia
  2. Para cada paper, manda o abstract pro Ollama (LLM rodando localmente) pedir uma tradução e resumo em português
  3. Formata a mensagem com título, autores, upvotes e o resumo traduzido
  4. Envia cada paper como mensagem separada no Telegram

O resultado no Telegram fica assim:

🤖 HuggingFace Papers — 27/05/2026
Top 8 papers do dia, traduzidos com llama3
──────────────────────────────

1. *Scaling Laws for Reward Model Overoptimization*
👥 Leo Gao, John Schulman, Jacob Hilton
💬 142 upvotes

📝 Investigamos como o desempenho de modelos de linguagem muda quando otimizamos 
excessivamente contra um modelo de recompensa (reward model). Descobrimos que o 
desempenho no proxy aumenta mas o desempenho verdadeiro diminui — um fenômeno 
conhecido como overoptimization. Nossos experimentos mostram leis de escala 
previsíveis para esse comportamento...

🔗 HuggingFace | arXiv


Pré-requisitos

  • Python 3.10+
  • Ollama instalado com llama3 (ou qualquer modelo que você tiver)
  • Um bot no Telegram (leva 2 minutos criar)

Estrutura do projeto

hf-digest/
├── hf_digest.py      # script principal
├── config.py         # suas configurações
└── get_chat_id.py    # helper para descobrir o chat_id


O código

config.py

Começa pela configuração. Tudo em um lugar só, fácil de ajustar:

TELEGRAM_BOT_TOKEN = "SEU_TOKEN_AQUI"
TELEGRAM_CHAT_ID   = "SEU_CHAT_ID_AQUI"
OLLAMA_URL         = "http://localhost:11434"
OLLAMA_MODEL       = "llama3"
MAX_PAPERS         = 8

Buscando os papers

A API do HuggingFace retorna os papers do dia em JSON. Simples assim:

def fetch_papers() -> list[dict]:
    today = datetime.now().strftime("%Y-%m-%d")
    url = f"https://huggingface.co/api/daily_papers?date={today}"

    headers = {"User-Agent": "Mozilla/5.0 (compatible; HFDigestBot/1.0)"}
    response = requests.get(url, headers=headers, timeout=20)
    response.raise_for_status()

    papers = response.json()
    return papers[:MAX_PAPERS]

Traduzindo com Ollama

Aqui mora a mágica. O Ollama expõe uma API REST local no localhost:11434. A gente manda um prompt e ele retorna a resposta do modelo:

def translate_with_ollama(text: str) -> str:
    prompt = f"""Você é um assistente especializado em IA e ML. 
Traduza e resuma o seguinte abstract para o português brasileiro.
Mantenha termos técnicos em inglês entre parênteses quando necessário.
Responda APENAS com o resumo, sem frases introdutórias.

Abstract:
{text}"""

    payload = {
        "model": OLLAMA_MODEL,
        "prompt": prompt,
        "stream": False,
    }

    response = requests.post(
        f"{OLLAMA_URL}/api/generate",
        json=payload,
        timeout=120,
    )
    return response.json()["response"].strip()

O stream: False faz o Ollama esperar processar tudo antes de responder — mais fácil de lidar do que streaming para esse caso de uso.

Enviando pro Telegram

O bot do Telegram aceita Markdown, então dá pra formatar bem as mensagens:

def send_telegram(text: str) -> bool:
    url = f"https://api.telegram.org/bot{TELEGRAM_BOT_TOKEN}/sendMessage"
    payload = {
        "chat_id": TELEGRAM_CHAT_ID,
        "text": text,
        "parse_mode": "Markdown",
        "disable_web_page_preview": True,
    }
    response = requests.post(url, json=payload, timeout=15)
    return response.ok

Juntando tudo no main()

def main():
    papers = fetch_papers()

    # Cabeçalho
    send_telegram(f"🤖 *HuggingFace Papers — {datetime.now().strftime('%d/%m/%Y')}*")

    for i, paper in enumerate(papers, start=1):
        abstract = paper["paper"].get("summary", "")

        translated = translate_with_ollama(abstract)
        message = format_paper(paper, i, translated)
        send_telegram(message)

    send_telegram(f"{len(papers)} papers enviados!")


Setup em 5 minutos

1. Instale a dependência

pip install requests

2. Crie o bot no Telegram

  • Abra o Telegram e procure @botfather
  • Envie /newbot, siga as instruções
  • Copie o token e cole em config.py

3. Descubra seu Chat ID

Envie qualquer mensagem pro seu novo bot e rode:

# get_chat_id.py
import requests
from config import TELEGRAM_BOT_TOKEN

r = requests.get(f"https://api.telegram.org/bot{TELEGRAM_BOT_TOKEN}/getUpdates")
updates = r.json()["result"]
chat_id = updates[-1]["message"]["chat"]["id"]
print(f"Seu chat_id: {chat_id}")

4. Suba o Ollama e rode

ollama serve
python3 hf_digest.py


Por que rodar o LLM localmente?

Poderia ter usado a API da OpenAI ou do Gemini para traduzir. Mas há algumas vantagens em rodar local:

  • Custo zero — sem pagar por token
  • Privacidade — os abstracts não saem da sua máquina
  • Sem rate limit — processa quantos papers quiser
  • Funciona offline — depois de baixar o modelo, não precisa de internet pra traduzir

O llama3 faz um trabalho muito bom em traduções técnicas. Se quiser mais velocidade, o phi4 ou gemma3 também funcionam bem e são mais leves.


Automatizando com cron

Para receber o digest todo dia de manhã sem precisar rodar manualmente, adicione ao crontab:

crontab -e

0 9 * * * cd /caminho/para/hf-digest && python3 hf_digest.py >> digest.log 2>&1


Possíveis melhorias

Algumas ideias para evoluir o projeto:

  • Filtrar por área: só receber papers de NLP, ou de RL, por exemplo
  • Score de relevância: pedir pro LLM avaliar o quão relevante é pra você com base no seu perfil
  • Salvar em banco: guardar um histórico local dos papers lidos
  • Interface web: um painel simples para ler os papers traduzidos no navegador
  • Suporte a outros feeds: conectar com arXiv diretamente ou com Papers With Code

Código completo

O projeto completo está disponível no GitHub: [https://github.com/lromais/ai_newsletters]


Se você testar e fizer alguma melhoria, compartilha nos comentários! E se tiver dúvida em algum passo do setup, é só perguntar.


Tags: python, ai, ollama, telegram, machinelearning