Welcome
My name is Siarhei Berdachuk.
I'm a seasoned IT professional with over 30 years of experience in software engineering, team leadership, and cutting-edge solution design. I specialize in Java development, AI integration with Spring AI, and high-performance backend systems.
This blog is where I share insights on clean code, test-driven development, scalable architectures, and practical applications of AI in enterprise environments.
Latest Articles
17-Sept-2026
Multimodal LLMs vs Dedicated ASR for Russian Speech: 9 Engines on One 16GB GPU
Russian ASR WER benchmark on an RTX 5060 Ti: GigaAM and faster-whisper beat multimodal LLMs on spontaneous speech; Gemma 4 12B QAT ties them on studio reads.
12-Sept-2026
Local AI Code Review with OpenCodeReview, OpenCode, and Qwen3.6-35B-A3B
A practical, local-first workflow for running deterministic AI code reviews through OpenCodeReview against an OpenAI-compatible Qwen3.6-35B-A3B endpoint.
09-Sept-2026
Qwen3.8-27B-NVFP4 on a 24GB RTX 4090: What a Dense Model Can and Cannot Do
Qwen3.8-27B-NVFP4 on RTX 4090 with FreeToken: dense weights fit, but the KV ceiling is ~5.6K tokens — 38/40 qualbench, 51/53 agentbench.
03-Sept-2026
Qwen3.6-35B-A3B on a 16 GB RTX 5060 Ti: Why Context Beats Tok/s for Coding
Qwen3.6-35B-A3B on a 16 GB RTX 5060 Ti with FreeToken: long-context tuning, FP8 vs NVFP4 throughput and correctness, and real memory requirements.