Brandon Bennett
116 posts
Brandon Bennett
@branben_
nyc 🗽 🎹 Music Ed 5+ YoE Forward Deployed Engineer - building “sound royale” - contributing to Omniroute (5.5K 🌟) always learning and open to chat
New York, NY Присоединился Şubat 2025
195 Подписки4 Подписчики
@kepano @flowybin 😂😂😂 stop stressing out my goat #obsidianheirarchy why do people hate searching for information it’s beyond me
English
on a budget or very tight budget and want to have your hermes agent run free models and rotate free keys automatically?
Shaf built a lightweight proxy that sits between Hermes and your AI providers... so when one free tier hits a rate limit, it automatically falls through to the next provider and rotates keys. No downtime, no manual switching
try it.. play with it.. let me know! git in the comments
English
@TheViableEdge @BzNav_R_Mont You should try @orca_build for this, Hermes isn’t native to orca but I’m working on the implementation now with a workflow similar to this! @JinjingLiang
English
@BzNav_R_Mont Fable tokens... no but really, it works best when Hermes is focued on project, its hard to jump from project to project with Hermes, so for rapid fire project hopping I'm in the Claude Code sessions.
English
Everyone shares their Hermes setup. Almost nobody shares their Hermes OPERATION.
I run a consulting + build practice (and a small production SaaS) with Hermes orchestrating Claude Code.
Here is the actual working model, not the installer:
English
agents consume data, and I need quick readable knowledge. generated a html just breaking down a suspicion I had on hermes tui in @orca_build @jinwoohong_ thinking long horizon here how over consumption of data might look
English
I added a tiny "papercuts" cli tool that agents can use to complain about bullshit they encountered during work, like dead-end tool calls, broken links, or other frustrations. The models would usually just push through without mentioning any problems
English
one thing I learned while being an educator is sometimes students can improve without communicating. thought about that while looking into an issue with @jinwoohong_ hermes preferred a problem solving pattern in @mattpocockuk wayfinder 😏 maybe i'll give them a backpack (folder)
English
we're hiring! both product and forward deployed engineers
mintlify.com/careers
Justin Torre@justinstorre
I am building a new team at Mintlify, and I am calling for strong engineers to come join me!! We are looking for folks that want to bring new enterprise product 0->1 at Mintlify. Strong builders, and ex-founders are perfect for this role. 👌
English
would love a new desktop agent super app
- lets me switch harnesses and models and multiplex across them
- makes it easy to move memory and context
- can orchestrate between models ( use Fable as a planner but a lower cost model for daily driver )
- can retroactively look at usage and optimize for cost / better results
English
@rauchg Something about the term “arena” makes it feel fun and competitive
English
Leader agent with two adversarial agents. Omniroute can route Claude and codex to other providers. Orca links to repo, user starts issue, this is the step I want to dive deeper. How much control can I give orca really?
English
working on improving how tasks are launched in @orca_build - goal: improve automations accessibility, project view. Today I asked myself “how can I make automations more appealing?” new workspc can store prompt but Hermes never handshakes and waits forever
English
@TencentHunyuan Hy3 is so good, I wish it could stay built into Hermes forever
English
@mattpocockuk wouldn’t the end goal make digesting code easier for everyone not just a select few
English
This is crazy to me
AI is an incredible lever for making PR's and commits easy to review
You just need a bit of steering to customize it to how you want
Kenton Varda@KentonVarda
I just declared a moratorium against AI-written change descriptions (e.g. PR and commit messages, also issues/tickets) from my team. AI was writing change descriptions that were worse than useless to me as I tried to review PRs: outlining details of the code that could easily be seen by looking at the code, but omitting the higher-level framing needed to understand broadly what the code is doing. I think people like having AI write these things because the output looks structured and thorough, which makes it feel professional in a way. But this isn't actually valuable. Concise, high-level descriptions are better for everyone. If I need to use my own AI to interpret what your AI wrote then something is wrong. Let AI write code, sure, but for the description, I'd rather see your prompt than your output. We could maybe have extended agents.md with guidelines on writing descriptions, but this seemed a bit pointless since a good, concise change description only takes a few minutes to write -- not a significant time savings to delegate to AI. At least, it doesn't take long if you understand the code -- and if you don't understand the code, then I'm definitely not merging it.
English
Merged PR #895 to model context protocol rust SDK today
relaxed tool output schema validation to full JSON schema.
backward-compatible
github.com/modelcontextpr…
English
I just merged a PR into the MCP Rust SDK!
getting hands on in the AI-agent ecosystem was a goal for me, and this PR represents one of those goals!
process?
read a spec, plan deeply, then ship safely.
shipping in v.3.0.0.
every downstream Rust MCP server
English
