roach

Sotfware engineer

[roach @dev_roach]

@dotey You can use this one

https://t.co/xApDAv3LHh

[roach @dev_roach]

@TaXue2025 @dotey You can use this one

https://t.co/xApDAv3LHh

[roach @dev_roach]

@real_kai42 What are different between kimi-code and kimi-cli?

[George from 🕹prodmgmt.world @nurijanian]

my favorite engineering skills for AI:

- Compound Engineering: https://t.co/BM7tA2RAHf
- Ryan Singer's shaping skills: https://t.co/yaWg0nI7Vm
- Matt Pocock's skills: https://t.co/0WtRqce6x5

I switched from Superpowers to Compound Engineering as they perfected the plugin over time, and I'm pretty sure I still only use like 10% of it

[Garry Tan @garrytan]

This sounds complicated but the agents can implement this in OpenClaw/Hermes Agent trivially (use skillify from GBrain with a link to this tweet)

Sounds ridiculous but you should try it

[Vaayne @LiuVaayne]

这是 yan5xu（ex ManusAI）写的长文：《从 Prompt 到 Harness：如何理解 LLM Engineering》

核心框架：螺旋上升的工程范式演进

第一圈：Prompt Engineering（2022-2024）

• 写好 system prompt，设计 few-shot，调 prompt 结构
• 天花板：模型变强后，精心设计的技巧反而成约束（Khan 的 Prompting Inversion：GPT-4o 上 Sculpting 97%，GPT-5 上降到 94%）

第二圈：Context Engineering（2025）

• Karpathy 命名："the delicate art and science of filling the context window with just the right information for the next step"
• 管理动态展开任务中的信息流转：上一步结果怎么流入下一步、环境反馈怎么注入、对话历史怎么压缩
• 典型产品：Cursor（动态检索相关代码）、Lovable/Bolt（LSP 报错、测试失败自动注入下一轮）
• 局限：规则是人硬编码的，"context rot"（多轮后性能平均下降 39%）

第三圈：Harness Engineering（2026-）

• 两步走：
‣ 放手：给 agent 工具链（lint、测试、搜索），让它自己决定什么时候用什么
‣ 上保险：框定能力边界（sandbox、CI 轮数限制、文件权限、结构测试）
• 定量验证：同一模型同一 prompt，换 harness 配置成功率从 42% → 78%

金句："Agents aren't hard; the Harness is hard."

关键洞察

• 不是线性替代，是螺旋积层：harness 内部还在跑 context engineering pipeline，pipeline 里还在写精心设计的 prompt
• 每一圈演化的驱动力：上一阶段不够用了 → 新实践被逼出来
• 下一圈瓶颈已露头：Eval（怎么定义"好"？LLM-as-judge 有 bias）和 Governance（多 agent 互认证、审计、权限）

[roach @dev_roach]

@aguming_ 특히나 컴팩션 부분에서 unicode 제대로 처리안하는 오픈소스 정말 많습니다

[roach @dev_roach]

@jietang Is it possible to use in coding plan?

[roach @dev_roach]

https://t.co/Z7pekMphk2

[roach @dev_roach]

Hermes Agent can now use Codegraph

CodeGraph gives AI coding agents a local semantic code index, so they can search symbols, trace callers/calles, and reason about impact without repeatedly scanning files.

I've contributed this features :)

@Teknium @NousResearch

https://t.co/Mxuu8BYkZS

[Tongyi Lab @Ali_TongyiLab]

1/6 Introducing Qwen3.5-LiveTranslate: Next-gen real-time interpretation is here. 🌍

We’re breaking down language barriers with 3,500+ language pairs, ultra-low latency, visual context, real-time voice cloning, and hotword customization.

Engineered to help you ship native, frictionless real-time translation experiences to a global audience.

[roach @dev_roach]

@9hills Is this pi?

[Adithya S K @adithya_s_k]

Introducing RL Environment Creator Skill
Now any one can create RL environments

$ npx skills add adithya-s-k/RL_Envs_101

> You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ...
> the repo has live working examples of environments that your coding agent can reference
> The skill is design to first understand what type of model you are training and create an environment while keeping that in mind

ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks.

Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.

[roach @dev_roach]

@JqOnly @realsigridjin @NousResearch @Teknium Thank you!

[roach @dev_roach]

I'm building an Ultragoal plugin for Hermes Agent after speaking at the first Hermes meetup in Seoul!

I will be release soon!

Thank you for having me @realsigridjin

Thank you for your support @NousResearch @Teknium

[roach @dev_roach]

Hermes Agent is my daily driver for coding

[roach @dev_roach]

@desilvakdn @realsigridjin @NousResearch @Teknium Thanks!

[Teknium 🪽 @Teknium]

Time for another Jam Session in the Nous Research discord on Hermes Agent!

This one focused on all the things you all are building. Come, bring the project(s) you are working on or join in to see all the other things people are using Hermes Agent for!

Tuesday May 12th, 4PM EST! See you then!

[roach @dev_roach]

@micheltamanda @Im_IrushiK In normal user case, you don't need to sub ubuntu pro

[roach @dev_roach]

I will be -> It will be 😂