Urgent.News

What's breaking now, across thousands of outlets.

Sunday, 16 August 2026 Reading Room

AI

🤖 I Built 2 Telegram Bots with Qwen3.8-Max — and the Results Were Seriously Impressive

💬 Following up on the story about the release of Qwen3.8-Max , I finally tried it on real-world tasks. Specifically, for building AI consultants for text channels (messengers) in my favorite programming language — Go . Spoiler: it’s really good, especially for such a low price per 1M tokens ! 😍 As a result, I built 2 demo Telegram bots, where GPT-4.1-mini acts as the brains 👇 1️⃣ A bot for…

dev.to ·

The article discusses the author's experience using Qwen3.8-Max to build two AI-powered Telegram bots. The first bot is designed to assist customers in booking car repair appointments by gathering vehicle details, addressing issues, providing pricing information, and scheduling convenient visit times. The second bot caters to furniture companies by taking kitchen measurements, answering questions about customization options, calculating costs, delivering final estimates, and arranging a design session at the company office.

The author emphasizes the impressive performance of Qwen3.8-Max, noting that it completed the tasks autonomously, including questioning itself at each stage and adjusting its reasoning and code. The token usage for the bots was approximately 12.8 million, which is significantly lower than that of DeepSeek V4 Flash Latest, which consumed over 15 million tokens for similar output results.

The author also highlights the efficiency of the development process using Kodik, a next-generation AI IDE developed by ArchiTech AI, a local company. They recommend trying Kodik and praise its high-quality features amidst the ongoing account bans faced by Anthropic and OpenAI.

Additionally, the article hints at the upcoming release of Qwen3.8-27b, a local model anticipated to be released by Alibaba, which may pose a challenge to the dominance of big AI model vendors. The author invites readers to contact them via Telegram (t.me/koddr) for further discussions on designing, building, and launching AI solutions for various business needs. The author encourages readers to follow their updates on multiple platforms including Telegram, VK, TenChat, Dzen, Teletype, and vc.ru.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Don't Start With RAG: Lessons From Building an Automotive AI Pipeline

When building an AI product, it's tempting to start with the fashionable pieces. Vector database. RAG. Agents. Multimodal models.

  • Focus on reliable structured data before RAG, using OBD codes and vehicle metadata
  • Normalize messy inputs (photos, voice, descriptions) into consistent evidence format
  • Preserve provenance and confidence levels for each input to guide LLM reasoning

·dev.to

Is It Okay to Trust AI Financial Advisory Services? Recently, a large securities firm in Korea provided financial advisory services through an AI (Artificial Intelligence) program, recommending stocks to buy and sell. On the 14th, the Financial Services Commission (FSC) warned that it would take strict measures against illegal financial advisory services, including those provided through AI programs. The FSC stated that some securities firms have been providing financial advisory services through AI programs, recommending stocks to buy and sell based on machine learning algorithms. However, the FSC emphasized that these services are not registered with the commission and therefore are illegal. The FSC explained that only financial investment advisory firms registered with the FSC are allowed to provide financial advisory services. The warning comes as some AI financial advisory services have been aggressively marketing themselves, with one service recommending that users “open a stock trading account and buy stocks for just 10 more weeks.” The FSC stated that it will take strict measures against these illegal services, including shutting down websites and taking legal action. The FSC advised consumers to be cautious when using AI financial advisory services and to verify whether the service is registered with the FSC before using it. In addition, the FSC recommended that consumers do not provide personal financial information to unregistered financial advisory services. The FSC also warned that consumers who use illegal financial advisory services may face penalties, including fines and imprisonment. The FSC will continue to monitor AI financial advisory services and take action against those that are operating illegally. Meanwhile, some experts have raised concerns about the potential risks associated with AI financial advisory services. They argue that AI programs may not fully understand the complexities of the financial market and may provide inaccurate or biased advice. As the use of AI financial advisory services continues to grow, it is essential for consumers to be aware of the potential risks and to take steps to protect themselves. The FSC will continue to work to ensure that financial advisory services, including those provided through AI programs, are operating fairly and transparently. In order to prevent such illegal cases, the FSC will strengthen its monitoring and inspection of financial advisory services. The FSC aims to protect consumers and maintain the integrity of the financial market.

·hani.co.kr

How I'm Learning AI in Public: My Roadmap

When I decided that I wanted to seriously start learning Artificial Intelligence, I quickly realized that one of the hardest parts wasn't finding resources. It was figuring out where to start.

·dev.to

More from Sunday 16 August →