Practice

This series of posts will be my way of documenting the journey of creating one of our team’s most ambitious products: the intelligent assistant, A.V.E.L.I.N. To give you some context, my development team and I are currently beta-testing the project within our Mozgii Ecosystem AI platform. Our primary focus is on A.V.E.L.I.N.—an intelligent personal assistant in Telegram built to handle both basic and complex tasks involving AI-powered search, processing, and analysis of information. ...

In traditional companies, everything is built on a clear hierarchy: decisions are made at the top and executed at the bottom. This approach might work in a stable environment, but for an IT startup, especially in AI, it stifles growth. IT companies need adhocracy: a management model where competence and results are valued more than titles. It’s about flexibility over bureaucracy and speed over approvals. The value of an idea is judged by its effectiveness, not by the position of its author. ...

Messengers have long evolved past being simple tools for exchanging messages. Today, they are the space where our work, personal lives, news, and friends converge. The logical next step is to integrate an AI assistant into this environment—one that helps structure communication and interacts with you in a familiar interface, just like any other contact. A majority of the solutions on the market are bots running on simplified or free versions of GPT. They can generate answers, but often fail to provide high-quality, deep information. ...

Users see a chat window. Sometimes voice, sometimes images. But behind this familiar interface lie radically different architectures and capabilities. Here are five key parameters that distinguish the top AI models in 2025: 1. Memory (Context Window) This defines how much information a model can retain within a single conversation. GPT-4o: 128k tokens (~300 pages of text) Claude 3 Opus & Gemini 2.5 Pro: Up to 1 million tokens (~2,000 pages) DeepSeek-VL Mini: ~8k tokens (~20 pages) More memory enables greater context and reduces hallucinations, but it also demands more powerful hardware. ...

Your Thought Process, Packaged Deep Research isn’t just another AI feature; it’s a fundamental shift toward an agent-based architecture. In this model, the LLM stops being a simple chatbot and becomes a co-author—an agent that independently searches, filters, validates, and structures information. What does this change? If you’re designing a business, a startup, or a product, you don’t have time to personally read 200 sources. Now, an AI agent does it for you. This frees you up to do the high-value work: to think, not just to search. ...

DeepSeek has once again followed its “quiet release” strategy, making its new DeepSeek-V3-0324 model available on Hugging Face without any major announcements. Instead of marketing hype, they’ve simply delivered a solution for the community to evaluate. I tested the model locally on a Mac Studio equipped with an M3 Ultra chip and saw impressive performance, generating over 20 tokens per second. This marks a significant acceleration for running capable models on local hardware, making it a viable option for developers. ...