I think this space is very untapped. Models are interesting, but I am absolutely obsessed with some things I've been researching/working on for the past few years:
Fractal tool discovery: tool taxonomy where an agent can "drill deeper" to find what specific tool it's looking for. Helps if/when polluting context with a zillion (mostly unnecessary) tools.
Leveraging splay trees: this is my favorite data structure and I think relatively unused in the context of agents/harnesses. A lot of times, recently-used workflows/tool-chains will be used again, so having those at the top of the search hierarchy is an awesome optimization.
Virtual containerized notebooks: models working in sandboxed (WASI) Python notebooks is incredible. Even local models (if given enough time) will usually converge on a good solution. Being able to mount tools/resources/fs is again, imo quite untapped. Some problems here are running native things (thing numpy/pandas) in containers is a nightmare (or impossible).
Anyway, happy to see other folks seriously doing stuff in this space. If anyone wants to collaborate on anything don't hesitate to reach out :) I'm also actively looking for a job or some contract gigs.
all other tools, mcps, apis, whatever are encapsulated by the one interface. new tools don't bloat the agent's context, and it can write its own code to perform more advanced and batch operations against the available tools (executed within a sandbox).
executor is a great implementation of this - <a href="https://executor.sh">https://executor.sh
opencode v2 also provides its own native implementation
dvt · · focus · HN ↗
Fractal tool discovery: tool taxonomy where an agent can "drill deeper" to find what specific tool it's looking for. Helps if/when polluting context with a zillion (mostly unnecessary) tools.
Leveraging splay trees: this is my favorite data structure and I think relatively unused in the context of agents/harnesses. A lot of times, recently-used workflows/tool-chains will be used again, so having those at the top of the search hierarchy is an awesome optimization.
Virtual containerized notebooks: models working in sandboxed (WASI) Python notebooks is incredible. Even local models (if given enough time) will usually converge on a good solution. Being able to mount tools/resources/fs is again, imo quite untapped. Some problems here are running native things (thing numpy/pandas) in containers is a nightmare (or impossible).
Anyway, happy to see other folks seriously doing stuff in this space. If anyone wants to collaborate on anything don't hesitate to reach out :) I'm also actively looking for a job or some contract gigs.
Fun times ahead.
kaurimu · · focus · HN ↗
mcp with 2 tools: execute(code) and search(query)
all other tools, mcps, apis, whatever are encapsulated by the one interface. new tools don't bloat the agent's context, and it can write its own code to perform more advanced and batch operations against the available tools (executed within a sandbox).
executor is a great implementation of this - <a href="https://executor.sh">https://executor.sh
opencode v2 also provides its own native implementation