"author:elifalmeida" matching MCP servers:

Turns a plain-language description of an LLM fine-tuning job into ranked hardware picks by estimating the required VRAM and FLOPs, then comparing live pricing across a dozen GPU clouds. Returns the cheapest, fastest, and most balanced configurations for a given budget or time limit, with each result labeled by its data source.

Enables AI agents and MCP clients to debug and observe LLM applications by querying traces, sessions, exceptions, prompts, datasets, scores, and metrics. Supports local stdio/HTTP access, selective tool loading, and read-only operation for safer observability workflows.