Unclaimed profileBuilt from public GitHub and Hugging Face activity. Matthew has not joined Lounge yet.Claim it
Not on Lounge yet
Matthew Charles Vladislav Busel
developerBuilding TensoRust, the evidence layer for AI. AI makes shit up. I make it cite sources. PhDs only.
Public projects
tokio-prompt-orchestrator
Tokio pipeline for serving LLM requests: bounded five-stage flow with dedup, circuit breakers, retries, rate limits and a dead-letter queue in front of Anthropic, OpenAI, llama.cpp or vLLM. REST, SSE, MCP, TUI.
Rustagentanthropicasync
Every-Other-Token
Rust CLI and web UI that intercepts an LLM token stream live: per-token confidence and perplexity, on-the-fly token mutation, provider and A/B prompt comparison, replay and heatmap export. OpenAI and Anthropic.
Rustanthropiccliinterpretability
LLM-Hallucination-Detection-Script
Rust CLI and library that flags low-confidence spans in LLM answers from token logprobs (offline or live via any OpenAI-compatible API) and renders per-token confidence as terminal, HTML or Markdown.
Rustai-safetyclihallucination-detection
FinRL_DeepSeek_Crypto_Trading
FinRL contest Task 1 entry: DQN ensemble (D3QN, DoubleDQN, TwinD3QN) trading 1-second BTC order-book data, with DeepSeek news sentiment and risk in the state. Research code.
Pythoncryptodeepseekfinance
llm-cpp
Single-header C++17 libraries for LLM features: OpenAI and Anthropic streaming, retries, caching, cost estimates, RAG, structured JSON output and tool-calling agents. Copy one .hpp, no SDK, no package manager.
HTMLaiai-agentsanthropic
Token-Visualizer
Free LLM token counter: see how GPT-4 and GPT-4o split your prompt into tokens, which lines cost the most, and which wordy phrases to cut, with savings measured by re-tokenizing. Windows .exe, macOS, Linux. Uses tiktoken.
Pythonchatgptcligpt-4
Likely stack
ClaudeOpenAILlamaPyTorchTransformersvLLMMCPFastAPIReact
