Projects with this topic
Sort by:
-
A 0.6-billion-parameter model that memorises an agent's tool catalog, with the data, decoder, serving and evaluation around it · TensorRT and TensorRT-LLM engineering, inference serving architecture, CUDA kernel development, MCP server design
Updated -
The MCP hub every agent in my estate connects to: memory and code intelligence, hosted tools, and a governed tool gateway · exactly-once delivery and crash recovery, MCP server design
Updated -
The execution substrate under everything else: one function definition becomes an orchestrated, budgeted, agent-callable tool · exactly-once delivery and crash recovery, tool synthesis and governed API integration, high-throughput batch inference, Po
Updated