シャドーイング練習: What Does an AI Engineer Do Now? - Aparna Dhinakaran, Arize AI - 動画で英語スピーキングを学ぶ
練習モードを選ぶ
各モードは全画面の練習ルームで開きます。
171 文
文が短すぎたり長すぎる場合は、Editをタップして調整してください。
1
Awesome.
2
Well, welcome to AI Engineer.
3
I thought given we're at AI Engineer New York, we'd kind of revisit the question of what does an AI engineer do?
4
I feel like every year we have an idea of what an AI engineer does, and that definition kind of changes.
5
And as I was putting this talk together, it was super timely,
6
because Karpathy actually just tweeted something this weekend where he kind of talks about this growing gap between
7
basically folks at the frontier who are running thousands of agents and swarms of always -on on agents,
8
and he really kind of drops this funnel where six billion plus users have maybe touched or interfaced with an LLM.
9
One to two billion are maybe like casual free tier users.
10
20 million are professionals who have maybe started using it for more, taking some piece of work that would have taken weeks down to days.
11
But really, those at the Frontier Labs are, which is a super small percentage, it's like 5 ,000 users,
12
are the ones who are running thousands and thousands of agents at once.
13
And as I look at this, it really is, I mean, he's kind of dropped a bit of the funnel.
14
I mean, the first question anyone looks at this is that, where am I on this?
15
And, you know, he's kind of drawn out a bit of a roadmap of how does everyone get to, in the next couple of years,
16
one of those 5 ,000 users who are running thousands of agents.
17
And this has been happening, I'd say, for a couple of decades now, where the abstraction layer has gotten higher and higher.
18
I mean, definitely faster in the last couple of years, but we've gone from engineers writing higher -level abstraction languages to,
19
I mean, just copy -pasting code, using local agents, to using cloud agents,
20
to really the expectation now is how do all of us start running a team of agents?
21
or factories of agents internally and build toward that.
22
And what this has led to is essentially what is the then and now of what an AI engineer is.
23
And this is a bit, I think, of a roadmap for some,
24
but I think internally this is what we've already started to kind of push towards.
25
As an AI engineer, your job is no longer to just write the feature, write an AI feature.
26
It's about how do you build the loop to write that feature.
27
It's no longer about just reading the diff and actually doing code review.
28
It's actually how do you set up systems and agents to do the code reviews,
29
and how do you write evals to actually do the non -deterministic, to review those non -deterministic tasks.
30
It's actually no longer even about just setting up metrics and software kind of telemetry, but it's actually how do you have agents go through
31
and read the observability data and and have agents actually consume that telemetry to find patterns.
32
And so this loop that we've kind of been talking about, which is the build, the eval, ship, you monitor, you improve it,
33
well, the AI engineer is now actually responsible, not just for doing each of those steps,
34
but how do you build the loop that actually does each of those steps?
35
And at Arise, one of the things we've rolled out in our product,
36
and I'm gonna talk about a bit of what we've learned in building this, is we built
37
Signal to actually run this loop for AI engineers.
38
What Signal does is that
39
Signal actually looks at all of the traces, it reads the telemetry, it identifies patterns in that telemetry,
40
and it can actually put up a fix or a PR if it has access to the repo.
41
And Signal, Signal's an always -on agent.
42
So this is an example of how
43
Signal starts, but it actually starts off by just reading the traces.
44
Super simple example, but it goes through, reads, and all this is configurable,
45
but a certain volume of traces that came in in some cadence,
46
it reads it, and then the goal is to actually identify issues from that telemetry data.
47
It has to keep track of issues that it's previously seen.
48
So we have to consider the memory of did it see something before
49
that it thought was an issue? did something seem like an issue
50
but it wasn't actually captured
51
and now as it collects more data it actually thinks it's much more of a critical critical issue
52
and so signal actually starts off by reading the traces to identify problems
53
and so one of the first things that that we realized was
54
that the future of who's actually consuming telemetry data has really changed
55
It's no longer about humans consuming telemetry data.
56
It's not even about humans consuming it and agents kind of taking on the task,
57
but it really is how do you set up your telemetry data so that agents are the primary consumers of that data?
58
And what that does is that that observability data can become the trigger for actually running and triggering these always -on agents.
59
And so the first thing that if you're building a team of, of a factory of really always on agents is,
60
what triggers these agents to wake up or start?
61
And there can be a variety of triggers.
62
For signal, for us, it's configurable to run on an ongoing cadence.
63
So every three hours, six hours, go pull the last X amount of traces, read them, identify problems.
64
It can also be triggered by events.
65
So Slack messages, and some user complains about something.
66
It can be a metric -based trigger, but these are essentially triggers to kick up the agent actually starting.
67
Once the agent actually starts, we have to spend some time figuring out, well, what does an agent run actually look like?
68
For us, we actually pull in all of the context of the traces, any additional third -party data,
69
so we allow collecting to APM logs, APM metrics, but all of the third -party telemetry data that we want to pull in,
70
access to the repo, access to the actual,
71
the access to any keys that are needed to actually run this agent,
72
but all of this information is put inside of a sandbox, and this is actually what's started up every time the agent wakes up.
73
And so the agent's triggered, we build a sandbox, We put everything that we need inside of that sandbox.
74
There's a proxy.
75
The only way to get information in and out of the sandbox is by the single proxy.
76
And everything that happens inside of that sandbox is actually already getting traced.
77
And every output, so the issues that are generated and created are all getting saved to some kind of memory.
78
And so this happens every single time the agent actually runs.
79
Every time the agent finishes running, so it's sweep through all of the traces, it either found something or it didn't.
80
There's what actually stays and what actually gets thrown away.
81
What stays is that if it found any issues, all of the output is actually saved off in some kind of memory
82
so that the next time an agent gets spun up, it doesn't have to start from ground zero.
83
It has a history of what are the issues that it actually found.
84
The agents are traced.
85
So all of the telemetry data of what the agent called, you know, if it actually put up a PR,
86
all of those things are actually traced.
87
And any skills that it may have learned while actually doing a run can also get saved to the actual repo.
88
But the things that go away is the sandbox gets torn down.
89
Any credentials, you know, obviously get removed.
90
The repo access is removed and so all of the stuff
91
that doesn't need to be saved kind of gets torn down after the run of Signal. And we've designed
92
Signal in a way to actually keep this agent durable.
93
So that means that either if the sandbox fails or if the harness fails,
94
then it shouldn't have to start everything from scratch in the middle of the run.
95
It should actually just be able to pick up from the step
96
that it failed and all of the state
97
so -called is kind of saved off in kind of a memory outside of the sandbox.
98
And so we've designed this to actually be a very durable agent so that if it fails, it can spin it back up.
99
You can launch a new sandbox, launch a new harness to kind of pick back up where it started.
100
So all of this is essentially the underpinnings of how we built signal to run every single run.
101
Agent gets triggered, agent gets started, loaded with all the context, the evidence, it sweeps through the telemetry data, it either finds something or it doesn't,
102
all of that gets saved to memory.
103
If something fails in the middle, the agent can get reset up.
104
And we built
105
Signal to also be agnostic of the model or the harness.
106
So today, Signal supports not just cloud code, which is what we started with, but it also supports cursor and codecs,
107
and the goal is to actually support all of the major, even the open source harnesses out there.
108
And Signal actually also supports multiple, will support kind of multiple models.
109
So even if you're using cloud code, you can pick any of the models inside the anthropic family.
110
And this is really important to us because,
111
one, it's.. an awesome way for us to evaluate the quality of the agent every time we spin it up.
112
It helps us control costs so we can see as we use more powerful models the better quality maybe the PRs are,
113
but can we reduce the quality of the, can we reduce the cost and actually still maintain the rigor
114
and maintain the quality of issues that are actually detected by signal?
115
So this is something we deeply thought about from first principles when we were designing signal.
116
And this is actually a pretty good segue into what does an AI engineer need to do.
117
Well, an AI engineer really needs to kind of learn how to write evals.
118
And so we actually, every time signal runs, signal finds an issue, it puts up a PR.
119
Well, there's evals that are run on top of the agent to test that PR.
120
And an eval can be as simple as testing something as part of your development workflow, you have a data set,
121
you can be a simple prompt with outputs that you're testing to make sure that there's no regressions.
122
But these evals can also, they need to be kind of aligned with human judgment.
123
And I'd argue
124
that this is actually probably where an AI engineer should be
125
spending a lot of their time as they're building the system to actually run the loop.
126
They should be actually spending time looking at the traces with their agents, identifying what the common errors are.
127
And there's a lot of skills.
128
I think I was hearing
129
Lance yesterday talk about the Claude skill to build evals.
130
We have a skill with Arise Evaluator to actually help you build and write evals, as well as align those evals with the human judgment.
131
but this is probably where the shift to defining what is good becomes really, really important for an AI engineer to actually do.
132
And for a lot of teams I think we speak with and we work with, their question becomes, well, how do I trust the evaluator?
133
And do I really trust that the eval is going to catch
134
or how do I trust that it's as good as my judgment?
135
And I'd say what we spend a lot of our own time doing for our own products is
136
that we actually evaluate the evaluators.
137
We spend a decent amount of time on actually building these data sets, golden fruit data sets.
138
We run evals, verify them against the human output, and then try to align to make sure
139
that the judge actually agrees with the human and they're in agreement before we actually ship out a new evaluator.
140
And these evals can then be used both in an offline testing as well as in an online testing manner.
141
And this is probably what I would say as an AI engineer is more and more of what the job is becoming.
142
And so what I'm going to leave you guys with is as an AI engineer, you've built this agent.
143
You've built this agent to actually run, be always on you run evals before you actually ship it into production.
144
You can ship it, and then once it goes into production, the goal is to actually trace everything.
145
And tracing everything now means not just tracing the agents, but also tracing the coding agents behind the agents that built this.
146
So you can trace cloud code, you can trace cursor, you can trace codecs, but all these coding agents can be traced
147
so you can understand what happened in the factory of agents that you all are building.
148
And then these agents that all the traces can then also be evaluated online.
149
So there's an offline evaluation when you're actually testing what a new change is,
150
but as it goes into production these become your online evals that you're running on top of the telemetry data.
151
This for us ends up becoming the new data that then gets pulled back into the agents to read the telemetry data,
152
the evals, and surface up kind of problems inside of the traces.
153
So the system really becomes, how do you go from collect data, eval that data,
154
run agents on top of it so that you can find problems
155
and figure out where the gaps are to actually improve your systems?
156
So if I go back to kind of the question that we started this whole thing with, It's what does an AI engineer do?
157
I mean the gap between,
158
I think there's a pretty big gap between what's happening in the labs today and what's happening with a lot of businesses.
159
And some of it could just be the data inside of
160
businesses are a lot more messy than what is very constrained and testable inside of a lab.
161
But AI engineers are kind of the role to close that gap.
162
And I think it really starts with tracing everything, tracing the agents you're actually building.
163
running evals, being able to learn how to test these products that we're really building, and then how do you build a system so
164
that agents can actually improve from the telemetry that's actually being collected so that you're no longer just building individual features,
165
but you're actually building the system to improve your products.
166
If you want to learn more, come check out.
167
We have a booth.
168
You can check out
169
Signal, learn more.
170
But thank you so much for listening to us.
171
Thank you.
アプリをダウンロード
流暢に話すために必要なすべて
AI発音採点文ごとに採点
IPA練習すべての音をマスター
単語学習自分の単語帳を作る
単語ゲーム遊びながら覚える
同じチャンネル
C11
Your AI Agent Starts Every Session Like a New Hire - Killian Carlsen-Phelan, Sonar
Daily ConversationUniversity StudentsWorking Adults
C15
The Multi-Agent Architecture That Actually Ships — Luke Alvoeiro, Factory
Business EnglishUniversity StudentsWorking Adults
C12
Anthropic's CCA Exam as a Field-Guide for Agentic Engineering — Frank Coyle, UC Berkeley
Business EnglishInterview EnglishUniversity Students
C1Why Software Factories Fail
Daily ConversationUniversity StudentsWorking Adults
C12
How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS
Business EnglishUniversity StudentsWorking Adults
C28
Beyond Code Coverage: Functionality Testing with Playwright — Marlene Mhangami, Microsoft
Business EnglishUniversity StudentsWorking Adults
C19
"Software Fundamentals Matter More Than Ever" — Matt Pocock
Business EnglishUniversity StudentsWorking Adults
C1Don't Build Agents, Build Skills Instead – Barry Zhang & Mahesh Murag, Anthropic
Business EnglishUniversity StudentsWorking Adults
FOR YOU
おすすめ動画
C12
70 文
A Practical Guide To Becoming An AI Engineer (2026)
Daily ConversationUniversity StudentsWorking Adults
C11
623 文
The Only SaaS Moats Left in 2026
Daily ConversationUniversity StudentsWorking Adults
C11
293 文
Is AI Killing Human Fashion Design? Who Will Shape the Future of Design? Fashion Tea #6
Daily ConversationUniversity StudentsWorking Adults
C12
195 文
Why Abu Dhabi Wants to Export Energy AI
Daily ConversationUniversity StudentsWorking Adults
C11
124 文
Agentic AI Frameworks Explained: Workflows, Multi-Agent, & Production
Daily ConversationUniversity StudentsWorking Adults
C11
510 文
Inside the secret race to build artificial intelligence | The Gray Area
Daily ConversationUniversity StudentsWorking Adults
C11
358 文
It’s not just you
Daily ConversationUniversity StudentsWorking Adults
C18
55 文
When AIs act emotional
Daily ConversationUniversity StudentsWorking Adults
C11
114 文
Harnessing generative AI in architecture, engineering & construction
Daily ConversationUniversity StudentsWorking Adults
C115
20 文
Alexandr Wang on what matters in an age of abundant intelligence
Daily ConversationUniversity StudentsWorking Adults
C13
148 文
900+ hours of Learning AI Agents in 9 Minutes
Daily ConversationUniversity StudentsWorking Adults
このレッスンの語彙とスピーキングのポイント
このスピーキングレッスンは、動画「What Does an AI Engineer Do Now? - Aparna Dhinakaran, Arize AI」を教材にしています。 繰り返し出てくる語は次のとおりです：agent, signal, eval, engineer, telemetry。 この動画には、シャドーイング用の文が171文、単語が2536語あります。 音声の長さは15:12です。 話す速さは1分あたり約167語で、日常会話に近い自然なペースです。 単語の86%は英語の頻出3,000語に含まれます。残りは練習の前に確認しておきましょう。
この動画の重要語彙
動画に出てくる覚えておきたい単語15語を、発音と意味つきで紹介します。
| 単語 | 発音 | 意味 |
|---|---|---|
| trace 名詞 | /tɹeɪs/ | 跡, 痕跡 |
| trigger 名詞 | /ˈtɹɪɡɚ/ | 引き金, 鉷 |
| evaluate 動詞 | /ɪˈvaljʊeɪt/ | 評する |
| loop 名詞 | /luːp/ | ループ |
| evaluation 名詞 | /ɪˌvæljuˈeɪʃn̩/ | 評価 |
| cloud 名詞 | /ˈklaʊ̯d/ | 雲, 雲霞 |
| output 名詞 | /ˈaʊtpʊt/ | 出力 |
| skill 名詞 | /skɪl/ | 腕, 技 |
| collect 動詞 | /ˈkɑlɪkt/ | 集める, 収集する |
| consume 動詞 | /kənˈsum/ | 消費する, 消耗する |
| spin 動詞 | /spɪn/ | 回転する |
| judgment 名詞 | /ˈd͡ʒʌd͡ʒ.mənt/ | 判決, 裁き |
| tear 動詞 | /tɛɹ/ | 破る, 引き裂く |
| frontier 名詞 | /ˈfɹʌntɪə/ | 境界, 国境 |
| sweep 動詞 | /swiːp/ | 掃除する, 掃く |
動画に出てくる句動詞
| 単語 | 発音 | 意味 |
|---|---|---|
| check out 動詞 | 試験する, 検査する | |
| wake up 動詞 | /ˌweɪk ˈʌp/ | 目覚める, 目を覚ます |
| go away 動詞 | 去る, 出る | |
| throw away 動詞 | 棄てる, 捨てる |
この動画の文法
話し手がよく使っている文型を、動画の実際の表現とともに紹介します。
| 文型 | 動画での表現 |
|---|---|
| 現在完了形 have/has + 過去分詞 — 過去の出来事が今も関係している | has gotten · we've gone · has led |
| 受動態 be + 過去分詞 — 誰がするかより、何が起きるかに焦点を当てる | wasn't actually captured · be triggered · are needed |
| 関係詞節 who / which + 節 — 人や物について情報を加える | are, which is · ones who are · about, which is |
注意したい発音
話し手はyou're, we've, doesn'tなど、短縮形や弱形を34回使っています。聞こえたとおりの短い形で発音しましょう。
- 「th」の音: booth /buθ/, anthropic /ænˈθɹɒp.ɪk/
- 「sh」と「zh」の音: evaluation /ɪˌvæljuˈeɪʃn̩/, essentially /ɪˈsɛnʃəli/, capture /ˈkæp.(t)ʃɚ/, casual /ˈkæʒ(u)əl/, expectation /ˌɛk.spɛkˈteɪ.ʃən/
- 長い単語（アクセントの位置に注意）: evaluation /ɪˌvæljuˈeɪʃn̩/, essentially /ɪˈsɛnʃəli/, internally /ɪnˈtɜːnəli/, expectation /ˌɛk.spɛkˈteɪ.ʃən/, telemetry /tɪˈlɛm.ɪ.tɹi/
日本語話者が苦手な音：
- /r/ と /l/ の区別 — /l/ は舌先を歯茎につけ、/r/ はどこにもつけない: durable /ˈdʊɹəbəl/, principle /ˈpɹɪn.sɪ.pəl/, telemetry /tɪˈlɛm.ɪ.tɹi/, configurable /kənˈfɪ.ɡə.ɹə.bəl/, credential /kɹɪˈdɛnʃəl/
- /v/ — /b/ にならないように、上の歯を下唇に当てる: evaluate /ɪˈvaljʊeɪt/, evaluation /ɪˌvæljuˈeɪʃn̩/, verify /ˈvɛɹɪfaɪ/, eval /ˈiː.væl/, revisit /ɹiːˈvɪzɪt/
- /f/ — 「フ」ではなく、上の歯と下唇で出す: frontier /ˈfɹʌntɪə/, offline /ɔfˈlaɪn/, define /dɪˈfaɪn/, folk /foʊk/, interface /ˈɪn(t)ɚˌfeɪs/
この動画での練習方法
- まず声を出さずに動画を最後まで聞き、知らない単語をメモします。
- まず0.75倍速で一文ずつシャドーイングし、慣れてきたら通常の速度に戻します。
- 自分の声を録音して元の音声と比べます。trace, trigger, evaluateなどの単語に特に注意しましょう。
シャドーイングとは？英語上達に効果的な理由
シャドーイング（Shadowing）は、もともとプロの通訳者養成プログラムで開発された言語学習法で、多言語習得者として知られるDr. Alexander Arguelles によって広く普及されました。方法はシンプルですが非常に効果的：ネイティブスピーカーの英語を聞きながら、1〜2秒の遅延で声に出してすぐに繰り返す——まるで「影（shadow）」のように話者を追いかけます。文法ドリルや受動的なリスニングと異なり、シャドーイングは脳と口の筋肉が同時にリアルタイムで英語を処理・再現することを強制します。研究により、発音精度、抑揚、リズム、連音、リスニング力、そして会話の流暢さが大幅に向上することが確認されています。IELTSスピーキング対策や自然な英語コミュニケーションを目指す方に特におすすめです。