- Published On
Part 1 of my code-first agent series: using agent-browser and CLI workflows for real research, verification, and structured web extraction.
Blog
Hands-on posts about LLMs, TTS, agents, benchmarks, and real-world AI security.
- Published On
A real-world React2Shell attack analysis: how fast unpatched Next.js apps were hit, what attackers executed, and what to rotate immediately.- Published On
Build low-latency streaming TTS with Orpheus, vLLM, and SNAC on a single RTX 3090, with notes on scaling the pipeline across GPUs.- Published On
How I used GRPO, Whisper WER, and a custom repetition penalty to improve LLaSA-style TTS quality with audio tokens.- Published On
Azure quietly removed Bing API resources. What changed, who it affects, and what the forced move to AI Agents Service means in practice.
Tags
llm (4)transformers (3)vllm (3)TTS (2)Text-to-Speech (2)LLM (2)quantization (2)awq (2)gptq (2)opensource (2)Hugging Face (2)Agents (1)CLI (1)Code-First (1)agent-browser (1)Automation (1)Codex (1)Vercel (1)MCP (1)Bash (1)Series (1)Security (1)Next.js (1)CVE (1)React2Shell (1)Cryptocurrency Mining (1)Orpheus (1)Kartoffel (1)vLLM (1)SGLang (1)Llasa (1)GRPO (1)Azure (1)Bing (1)API (1)Shutdown (1)tools (1)llama (1)local-llm-series (1)azure (1)openai (1)categorization (1)