@nijatzeynalov
AI Engineer
Nothing here yet.
Nothing here yet.
Aug 6, 2025 · 9 min read · We use AI every day. We ask it questions, we tell it to write stories. But have you ever wondered how it actually thinks? We know these LLMs are made of "neurons," digital versions of the ones in our own brains, but what do they do? Does one neuron s...
Join discussion
Jul 21, 2025 · 5 min read · LLMs have consumed a vast library of information from the internet. However, a critical limitation exists: their knowledge is frozen in time. LLM trained in 2023 would be oblivious to most world events, scientific discoveries, or new laws from 2024. ...
Join discussion
Jul 7, 2025 · 7 min read · We often face a classic dilemma: the most powerful models are also the most resource-hungry. A model like bert-base-uncased, with its 109 million parameters, can achieve best results but is often a non-starter for applications that demand low latency...
Join discussion
Jul 7, 2025 · 5 min read · Whisper is computationally intensive, and running it efficiently at scale or on constrained devices is a real engineering challenge. To make Whisper more suitable for production deployment, I explored two independent and hardware-aware optimization s...
Join discussion
Jul 7, 2025 · 5 min read · The Whisper model by OpenAI is a powerful multilingual, multitask ASR system. However, this generality can become a liability when adapting Whisper to a specific language and task. For example, transcribing Azerbaijani speech without translation or l...
Join discussion