Masaki.K | AI触りまくるEM (@Emukei_) header
Masaki.K | AI触りまくるEM (@Emukei_) profile picture

AIエージェントと毎日いっしょに働いてるEM。Claude Code / Devin / Grok Botあたりを触った感想を、ゆるくつぶやいてます。AIを使って稼ぐ実験のライブ配信します。茅ヶ崎住まい👶

茅ヶ崎
Joined November 2009
Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)Photo by Masaki.K | AI触りまくるEM (@Emukei_)
[固定用]はじめましての方へ。EMやってて、チームや組織にAIエージェント入れる係みたいになってます。 Claude Code / Codex / Devin / Grokbotあたりを毎日さわって、良かったところも罠だったところもゆるく書いてます。 個人開発でmacアプリ作ったり、AIツール触ったり、ずんだのショート作ったりも。 同じようなことやってる人、気軽に絡んでください〜
3
280
エージェントは「工場」で回すか、「チーム」として管理するか。 Zachさんは、人じゃないから測って直せる工場派らしい。 Devin Botsはチーム側の例だけど、在るべきが決まってる作業は工場っぽく回したくなる
After two years of rapid change in software development, we are again at a crossroads, with two competing visions for what the future might hold. Vision one is software factories (e.g. Warp Factories), code-defined, closed-loop systems for continuously measuring and improving cloud coding agents. Software factories are developer infrastructure. Platform teams do “factory engineering” on them to optimize development across an entire product, getting the most productivity from agents while minimizing cost and improving quality. It’s a system design mindset for deploying agents. Vision two is agent teams (e.g. Devin Bots). In this approach, individuals manage agents to orchestrate complex tasks. For example, you might have a “chief of staff” agent and ask that agent to do a coding project. It decomposes the project into subtasks that different agents take on (e.g. a designer, an implementer), reports on progress, and eventually ships. These teams can also have “always on” agents that proactively fix things, respond to events, and so on. Unlike with software factories, the user experience is more about “managing” agents, akin to how you might manage a person or a team. Which one is better depends on how you conceive of the problem. I view the problem as: how do you securely deploy coding agents to maximize product throughput and quality, while minimizing cost? I believe this is what businesses care about – there is this magical new technology for building software at scale and they want to make sure they are using it to get the biggest possible competitive advantage. Put simply: this is an optimization problem. We have a new engineering primitive, the coding agent, and we want to use that primitive as well as we can. Framing it in terms of optimization immediately leads to a need for measurement; after all, you can only optimize what you can measure. In order to get to measurement, you need reliable, standardized data. In order to get data, you need some amount of centralization and standardization, otherwise that data would be too noisy and heterogenous to do anything with. In order to centralize, you need to move to the cloud and define what a factory is in code. Your factory needs to be open and flexible so you have degrees of freedom for testing different models, harnesses and skills. You need ways of experimenting to see if different configurations perform better. And then you need to be able to feed those learnings back into the system, so it improves over time. When you frame the problem clearly, the shape of the solution becomes much more obvious. Compared to agent teams, the factory engineering approach is much better suited to solving the optimization problem because it puts measurement, repeatability, and standardization at the center. It allows for a principled approach to improvement in the long run that is missing with agent teams. You want your factory to scale like an engineered system, not like an increasingly large organization (don’t make your agents work in a big company!). The engineering approach is possible precisely because agents are not people; for the first time in the history of software development we can directly measure, grade and improve how code and products are shipped. We shouldn’t waste this opportunity by anthropomorphizing the technology and treating it like a teammate. Also, wouldn’t you rather engineer an agent system than manage a team of agents? I’ve been both a builder and a manager for most of my career. You can guess which I prefer.
62
TLA+とLeanという言葉を知ったので、学習📚決定論的チェックにどう活かせるのか
53
めっちゃ分かる。。開発組織の正式な見積もりとして、それを正としてしまうと、そこまで使えない人が受け持った時に大きく破綻するので、なかなかAIパワーを反映させづらい
最近困るのが、工数どんくらいですか? って質問 マジで読めなくね? 集中したら大抵1日で終わるし
1
109
Masaki.K | AI触りまくるEM retweeted
並行セッション立てて、次の指示を出しておく
仕事でClaude使ってる人、「待ち」の時間なにしてんの
2
4
12
3,679
本当にこれ。コスト感とかセキュリティ身につけたい人は、個人開発して何かデプロイしてサービス公開してみてほしい。 身銭を切って、明細をみれば怖さを知れば自ずと身に付く。。
コストとかセキュリティなんて、自分の身に降りかからないと人間本気にならないよね。個人開発大事
1
3
128
potetoさんのAll Code is Slop (Except Mine)の図解。下の層から足りてないのを整備していきたい。すでにコメント過多になってるところ思い当たっている
1
61
Devin Botsのおかげで、バージョンアップ作業、不具合修正など在るべきが決まってる作業からエンジニアが解放されるイメージが沸いてきた。Devin for Startupsでもらったクレジットもあるしやるぞー
52
Grok Botの自動化、「手順じゃなくて目標から渡す」がほんとそれ。 目標とやり方含めて、音声でダーっと話すことは最近のお気に入り。言い直させるのいいね。取り入れよう
I think a lot of people aren't using Grok Bot to its full potential, because knowing how to automate things is actually quite a difficult skill to learn. I wrote this post to break down how I approach automating anything with Grok Bot!
53
Masaki.K | AI触りまくるEM retweeted
Here's what @cognition is doing right now. They're giving away 10,000 Devin Max accounts. Each one is worth $200 and comes with up to $10,000 in credits. SWE-2 is free, and it's a great model built on Kimi K3. SWE-2 Lightning is multiple times faster than Sol Ultrafast at less than a tenth of the price. Devin Bots, done right. Enterprises get Mythos access for security work. And individuals like you and me get advanced security features you'd usually only find reserved for companies and enterprises. There's a support program that covers up to $65,000 in credits. Devin Cloud comes with Windows, Linux and Mac on top of your subscription. And the best models the moment they drop. Then there's the ambassador program. 6 months of Devin Max plus credits, and they don't ask you for anything or lock you into anything. It's completely up to you to share your experience however and whenever you see fit. They don't even require a follower count. I've seen fellow ambassadors without big numbers, but they're valuable people with real experience, and Cognition appreciates that. In all my time as an ambassador they've never once asked me to praise anything. Most of the negatives I called out in my Devin review got fixed over time, and they were never bothered by it. Trust me, in this space most companies get upset the moment you talk about their flaws. They care, they empathize, and they listen. That's what makes them the exception for me. And the most important part: a great team and great leaders who want you to grow. They help you get bigger, give you advice, and open doors to relationships and opportunities without expecting anything in return. They've given me more capacity to give to people and support them. Honestly, I've never been prouder to be a Cognition ambassador. I'll always be with them, and I'll do my best to be one of them and a bridge between them and the community.
55
14
1
225
12,775
Masaki.K | AI触りまくるEM retweeted
There’s been a growing volume of threats such as Java4Shell and NGINX2Root, just this week. To help, we’re launching Devin Security Threats: a managed index of new vulnerabilities disclosed, integrated with Devin for one-click scanning and remediation. devin.ai/blog/nginx2root-jav…
3
35
14
351
50,121
Teamアカウントを既に持っててログイン済みの時はログアウトしといた方が良い。個人freeのアカウントにMax付与したかったのに、間違ってTeamにいったみたいで消えてしまった。。問い合わせ中
We’re giving away up to $100M in Devin Bots this week! Just text Poke and convince them to build your idea: poke.com/bots
1
59
サポート問い合わせでわりとすぐに回答してもらえました! Teamには付与されないので、個人のメールアドレスみたいなアカウントを新たに作ると、そこにmaxのofferが出てきた。同じような人がいたら、とりあえずサポート問い合わせが良さそうです。
24
Teamでも最大500$までもらえるので、いくつかのリポジトリにClaude Managed Agentで指摘の学習基盤付きのコードレビューエージェント作って運用してます。Opus5.5にしてるけど、割といい感じの指摘くれます。
Claude のAPIクレジットについて、無料クレジットが毎月配布されているの意外と知られていないっぽい? Max: $100 Ultra: $200 Team: Standard $20 / Premium $100 無料でAPIクレジット使えるので、利用した方が良いです。 platform.claude.com/settings…
1
2
969
ま?Googleさん、どれだけモデルだすの。これが最上位モデルってことか
tomorrow DeepMind will be dropping Gemini 4 Sodium, a fast yet powerful frontier model which costs just $11.75/M (introductory pricing, raises to $16.50 soon) and is their best model for various things and stuff, rivaling Argon, Carbon, Hydrogen, and Neon in several situations
56
激アツ!弊社にもClaude Meetingsがロールアウトされた!会議中に「ねえClaudeくん」って話しかけたらチャットで回答してくれる新感覚。しばらく会議が楽しみだあ
1
56
Googleの動きが激しい。NeonがArgonの上位モデルってこと?元素でいくならヘリウムもどこかで出るのかね。上とか下とかイメージがつかない領域になってきた。。
Google continues to accelerate its release cadence with Gemini 4 Neon, delivering a small Intelligence gain over Gemini 4 Carbon at a ~17% lower Cost per Task @GoogleDeepMind has released Gemini 4 Neon, its fourth model in just seven weeks. Neon (xhigh) scores 61 on the Artificial Analysis Intelligence Index, 1 point ahead of Gemini 4 Carbon (high) and 3rd overall, behind GPT-6.2 Astra (max) and Claude Fable 5.2 (max). At high effort, Neon matches Carbon (high) at less than half the Cost per Task using Carbon’s standard (non-discounted) pricing. Key takeaways: ➤ Slight improvement in Intelligence for lower Cost and Time per Task: Neon (xhigh) scores 61 on the Intelligence Index, 1 point ahead of Carbon (high), with small improvements in agentic knowledge work, visual reasoning and coding. It costs $6.44 per task, ~17% less than Carbon ($7.76 at non-discounted pricing), and takes ~21 minutes per task, ~40% less than Carbon (~36 minutes). Neon uses ~15% more output tokens per task, so the lower cost comes purely from its 25% lower token prices ($3/$15 vs $4/$20 per 1M input/output tokens), and the lower time from its higher output speed (~110 vs ~57 output tokens per second) ➤ Competitive with Claude Opus 5.5 and Claude Fable 5.2 on coding agents: With Antigravity CLI, Neon (xhigh) scores 59 on the Artificial Analysis Coding Agent Index, 3 points ahead of Carbon (high, 55). It scores in line with Claude Code with Claude Fable 5.2 (xhigh, 58) and Claude Opus 5.5 (max, 58) at a much lower Cost per Task. Gemini 4 Neon in Antigravity has an average Cost per Task of $6.86, compared to $12.61 for Claude Opus 5.5 (max) and $15.47 for Claude Fable 5.2 (xhigh). Neon is available in Antigravity CLI for Pro and Ultra subscribers ➤ Significant improvements in visual reasoning and output presentation quality: In GDP.pdf, Gemini 4 Neon (high) scores 33.4%, 5 points above Carbon (high, 28.4%) and in line with the highest scores measured, while Neon (xhigh) scores 28.0%. On AA-Briefcase, Neon (xhigh) scores 1700 Elo overall and leads on Presentation Quality with 1771 Elo, 45 points ahead of the previous leader Claude Sonnet 5.5 (max, 1726). On Google’s Multimodal Knowledge Agent evaluation (MKA), Neon reaches a Climb WSR of 56%, up 2 points from Carbon and on par with the leader, Claude Fable 5.2 (max, 57%) Key model details: ➤ Context Window: 1M tokens ➤ Multimodality: Text, image, video, and speech input, with text output ➤ Availability: Gemini 4 Neon is publicly available at launch through Google’s API. Pro and Ultra subscribers can also access it in Antigravity and the Gemini app ➤ Effort levels: Gemini 4 Neon offers four effort levels (low, medium, high, and xhigh) - xhigh was introduced to Gemini models in the recent Hydrogen release ➤ Pricing: $3/$15 per 1M input/output tokens, 50% above Gemini 4 Carbon’s currently discounted price of $2/$10 (which ends on November 30), and 25% below its standard $4/$20. Cached input tokens receive the same 95% discount ($0.15)
55
Masaki.K | AI触りまくるEM retweeted
2022 was the year of autocomplete 2023 was the year of chat 2024 was the year of the AI IDE 2025 was the year of coding agents 2026 was the year of agent orchestration 2027 will be the year of outcome engineering
127
92
39
1,424
724,986
Devin Bot、少し触ってみた感想 ・QAとかProduct Engineerなど文脈単位で専門botを作り、それが人間の依頼をオーケストレーションして、セッションへ作業を投げてくれる ・セッション自体の使用量は本人のクオータから捻出される。つまり個人full seatから使われるのでコスパもよい ・専門botは課金対象、だけど仲介するだけなのでコストは大きくならなそう
Meet Devin Bots. From writing firmware to merging thousands of PRs a week, a Devin Bot leads a team of agents to get the job done.
1
146
Masaki.K | AI触りまくるEM retweeted
Devin Bots at Cognition have led to us to a pattern of extreme automation (Bots built Bots) A few very practical and powerful examples from my personal use:
Meet Devin Bots. From writing firmware to merging thousands of PRs a week, a Devin Bot leads a team of agents to get the job done.
11
16
1
199
15,084