Практика Shadowing: What Does an AI Engineer Do Now? - Aparna Dhinakaran, Arize AI - Изучайте разговорный английский по видео

Выберите режим практики

Каждый режим открывает полноэкранную комнату практики.

171 предложений

Если предложения слишком короткие или длинные, нажмите Edit, чтобы их изменить.

Awesome.

Well, welcome to AI Engineer.

I thought given we're at AI Engineer New York, we'd kind of revisit the question of what does an AI engineer do?

I feel like every year we have an idea of what an AI engineer does, and that definition kind of changes.

And as I was putting this talk together, it was super timely,

because Karpathy actually just tweeted something this weekend where he kind of talks about this growing gap between

basically folks at the frontier who are running thousands of agents and swarms of always -on on agents,

and he really kind of drops this funnel where six billion plus users have maybe touched or interfaced with an LLM.

One to two billion are maybe like casual free tier users.

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.

But really, those at the Frontier Labs are, which is a super small percentage, it's like 5 ,000 users,

are the ones who are running thousands and thousands of agents at once.

And as I look at this, it really is, I mean, he's kind of dropped a bit of the funnel.

I mean, the first question anyone looks at this is that, where am I on this?

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,

one of those 5 ,000 users who are running thousands of agents.

And this has been happening, I'd say, for a couple of decades now, where the abstraction layer has gotten higher and higher.

I mean, definitely faster in the last couple of years, but we've gone from engineers writing higher -level abstraction languages to,

I mean, just copy -pasting code, using local agents, to using cloud agents,

to really the expectation now is how do all of us start running a team of agents?

or factories of agents internally and build toward that.

And what this has led to is essentially what is the then and now of what an AI engineer is.

And this is a bit, I think, of a roadmap for some,

but I think internally this is what we've already started to kind of push towards.

As an AI engineer, your job is no longer to just write the feature, write an AI feature.

It's about how do you build the loop to write that feature.

It's no longer about just reading the diff and actually doing code review.

It's actually how do you set up systems and agents to do the code reviews,

and how do you write evals to actually do the non -deterministic, to review those non -deterministic tasks.

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

and read the observability data and and have agents actually consume that telemetry to find patterns.

And so this loop that we've kind of been talking about, which is the build, the eval, ship, you monitor, you improve it,

well, the AI engineer is now actually responsible, not just for doing each of those steps,

but how do you build the loop that actually does each of those steps?

And at Arise, one of the things we've rolled out in our product,

and I'm gonna talk about a bit of what we've learned in building this, is we built

Signal to actually run this loop for AI engineers.

What Signal does is that

Signal actually looks at all of the traces, it reads the telemetry, it identifies patterns in that telemetry,

and it can actually put up a fix or a PR if it has access to the repo.

And Signal, Signal's an always -on agent.

So this is an example of how

Signal starts, but it actually starts off by just reading the traces.

Super simple example, but it goes through, reads, and all this is configurable,

but a certain volume of traces that came in in some cadence,

it reads it, and then the goal is to actually identify issues from that telemetry data.

It has to keep track of issues that it's previously seen.

So we have to consider the memory of did it see something before

that it thought was an issue? did something seem like an issue

but it wasn't actually captured

and now as it collects more data it actually thinks it's much more of a critical critical issue

and so signal actually starts off by reading the traces to identify problems

and so one of the first things that that we realized was

that the future of who's actually consuming telemetry data has really changed

It's no longer about humans consuming telemetry data.

It's not even about humans consuming it and agents kind of taking on the task,

but it really is how do you set up your telemetry data so that agents are the primary consumers of that data?

And what that does is that that observability data can become the trigger for actually running and triggering these always -on agents.

And so the first thing that if you're building a team of, of a factory of really always on agents is,

what triggers these agents to wake up or start?

And there can be a variety of triggers.

For signal, for us, it's configurable to run on an ongoing cadence.

So every three hours, six hours, go pull the last X amount of traces, read them, identify problems.

It can also be triggered by events.

So Slack messages, and some user complains about something.

It can be a metric -based trigger, but these are essentially triggers to kick up the agent actually starting.

Once the agent actually starts, we have to spend some time figuring out, well, what does an agent run actually look like?

For us, we actually pull in all of the context of the traces, any additional third -party data,

so we allow collecting to APM logs, APM metrics, but all of the third -party telemetry data that we want to pull in,

access to the repo, access to the actual,

the access to any keys that are needed to actually run this agent,

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.

And so the agent's triggered, we build a sandbox, We put everything that we need inside of that sandbox.

There's a proxy.

The only way to get information in and out of the sandbox is by the single proxy.

And everything that happens inside of that sandbox is actually already getting traced.

And every output, so the issues that are generated and created are all getting saved to some kind of memory.

And so this happens every single time the agent actually runs.

Every time the agent finishes running, so it's sweep through all of the traces, it either found something or it didn't.

There's what actually stays and what actually gets thrown away.

What stays is that if it found any issues, all of the output is actually saved off in some kind of memory

so that the next time an agent gets spun up, it doesn't have to start from ground zero.

It has a history of what are the issues that it actually found.

The agents are traced.

So all of the telemetry data of what the agent called, you know, if it actually put up a PR,

all of those things are actually traced.

And any skills that it may have learned while actually doing a run can also get saved to the actual repo.

But the things that go away is the sandbox gets torn down.

Any credentials, you know, obviously get removed.

The repo access is removed and so all of the stuff

that doesn't need to be saved kind of gets torn down after the run of Signal. And we've designed

Signal in a way to actually keep this agent durable.

So that means that either if the sandbox fails or if the harness fails,

then it shouldn't have to start everything from scratch in the middle of the run.

It should actually just be able to pick up from the step

that it failed and all of the state

so -called is kind of saved off in kind of a memory outside of the sandbox.

And so we've designed this to actually be a very durable agent so that if it fails, it can spin it back up.

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.

Скачать приложение

Всё для свободной речи

ИИ-оценка произношенияОценка каждой фразы

Практика IPAОсвойте каждый звук

СловарьСоберите свой словарный запас

Игра со словамиУчитесь играя

Сканировать QR

Download on theApp Store Get it onGoogle Play

Тот же канал

Your AI Agent Starts Every Session Like a New Hire - Killian Carlsen-Phelan, Sonar

Daily ConversationUniversity StudentsWorking Adults

The Multi-Agent Architecture That Actually Ships — Luke Alvoeiro, Factory

Business EnglishUniversity StudentsWorking Adults

Anthropic's CCA Exam as a Field-Guide for Agentic Engineering — Frank Coyle, UC Berkeley

Business EnglishInterview EnglishUniversity Students

Why Software Factories Fail

Daily ConversationUniversity StudentsWorking Adults

How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS

Business EnglishUniversity StudentsWorking Adults

Beyond Code Coverage: Functionality Testing with Playwright — Marlene Mhangami, Microsoft

Business EnglishUniversity StudentsWorking Adults

"Software Fundamentals Matter More Than Ever" — Matt Pocock

Business EnglishUniversity StudentsWorking Adults

Don't Build Agents, Build Skills Instead – Barry Zhang & Mahesh Murag, Anthropic

Business EnglishUniversity StudentsWorking Adults

FOR YOU

Лексика и советы по произношению к этому уроку

Этот урок разговорной практики построен на видео «What Does an AI Engineer Do Now? - Aparna Dhinakaran, Arize AI». Чаще всего повторяются слова: agent, signal, eval, engineer, telemetry. В этом видео 171 предложений и 2536 слов для шедоуинга. Речь длится 15:12. Говорящий говорит в естественном темпе, около 167 слов в минуту, близком к повседневной речи. 86% слов входят в 3000 самых частых слов английского языка; остальные стоит разобрать перед занятием.

Ключевая лексика этого видео

15 слов из видео, которые стоит выучить, с произношением и значением:

Слово	Произношение	Значение
trace существительное	/tɹeɪs/	слежка, прослушка
trigger существительное	/ˈtɹɪɡɚ/	спусково́й крючо́к, куро́к
evaluate глагол	/ɪˈvaljʊeɪt/	оце́нивать, оцени́ть
loop существительное	/luːp/	пе́тля, петля́
evaluation существительное	/ɪˌvæljuˈeɪʃn̩/	оце́нка, оце́нивание
cloud существительное	/ˈklaʊ̯d/	о́блако, ту́ча
output существительное	/ˈaʊtpʊt/	произво́дство, проду́кция
skill существительное	/skɪl/	уме́ние, на́вык
collect глагол	/ˈkɑlɪkt/	собира́ть, собра́ть
consume глагол	/kənˈsum/	потреблять, расходовать
harness глагол	/ˈhɑɹ.nəs/	запряга́ть
essentially наречие	/ɪˈsɛnʃəli/	по существу́, суще́ственно
spin глагол	/spɪn/	верте́ться, крути́ться
judgment существительное	/ˈd͡ʒʌd͡ʒ.mənt/	осужде́ние, сужде́ние
metric прилагательное	/ˈmɛt.ɹɪk/	метри́ческий

Фразовые глаголы, которые вы услышите

Слово	Произношение	Значение
put up глагол		ста́вить
set up глагол	/ˌsɛt ˈʌp/	устана́вливать, установи́ть
check out глагол		проверя́ть, прове́рить
figure out глагол		разбира́ться, разобра́ться
wake up глагол	/ˌweɪk ˈʌp/	просыпа́ться, просну́ться
back up глагол		дава́ть за́дний ход
end up глагол		зака́нчиваться, зако́нчиться
go away глагол		уходи́ть, уйти́

Грамматика в этом видео

Конструкции, которые говорящий использует чаще всего, с точными словами из видео:

Конструкция	В видео
Present Perfect have/has + причастие прошедшего времени — прошлое действие, важное сейчас	has gotten · we've gone · has led
Пассивный залог be + причастие прошедшего времени — важно, что происходит, а не кто это делает	wasn't actually captured · be triggered · are needed
Придаточные определительные who / which + предложение — уточнение о человеке или предмете	are, which is · ones who are · about, which is

Произношение, на которое стоит обратить внимание

Говорящий использует 34 сокращённых и редуцированных форм, например you're, we've, doesn't. Произносите их коротко, так, как слышите.

Звуки «th»: booth /buθ/, anthropic /ænˈθɹɒp.ɪk/
Звуки «sh» и «zh»: evaluation /ɪˌvæljuˈeɪʃn̩/, essentially /ɪˈsɛnʃəli/, capture /ˈkæp.(t)ʃɚ/, casual /ˈkæʒ(u)əl/, expectation /ˌɛk.spɛkˈteɪ.ʃən/
Длинные слова — следите за ударением: evaluation /ɪˌvæljuˈeɪʃn̩/, essentially /ɪˈsɛnʃəli/, internally /ɪnˈtɜːnəli/, expectation /ˌɛk.spɛkˈteɪ.ʃən/, telemetry /tɪˈlɛm.ɪ.tɹi/

Звуки, трудные для русскоязычных:

/w/ — губы округлены, это не /v/: sweep /swiːp/, tweet /ˈtwiːt/, segue /ˈsɛɡweɪ/, swarm /swɔɹm/, workflow /ˈwɝkfloʊ/
Звонкие согласные в конце слова — не оглушайте их: cloud /ˈklaʊ̯d/, lab /læːb/, percentage /pə(ɹ)ˈsɛntɪd͡ʒ/, log /lɔɡ/

Как заниматься с этим видео

Прослушайте всё видео один раз молча и выпишите незнакомые слова.
Начните со скорости 0,75×, повторяйте предложение за предложением и вернитесь к обычной скорости, когда станет легко.
Запишите себя и сравните с оригиналом, обращая внимание на такие слова, как trace, trigger, evaluate.

Что такое техника Shadowing?

Shadowing — это научно обоснованная техника изучения языка, изначально разработанная для подготовки профессиональных переводчиков и популяризированная полиглотом доктором Александром Аргуэльесом. Метод прост, но эффективен: вы слушаете аудио на английском от носителей языка и немедленно повторяете вслух — как тень, следующая за говорящим с задержкой в 1–2 секунды. В отличие от пассивного прослушивания или грамматических упражнений, Shadowing заставляет мозг и мышцы рта одновременно обрабатывать и воспроизводить реальные речевые паттерны. Исследования показывают, что это значительно улучшает точность произношения, интонацию, ритм, связную речь, понимание на слух и беглость речи — что делает его одним из самых эффективных методов для подготовки к IELTS Speaking и реального общения на английском.

Техника шедоуинга: читать полное пошаговое руководство →

Практика Shadowing: What Does an AI Engineer Do Now? - Aparna Dhinakaran, Arize AI - Изучайте разговорный английский по видео

Выберите режим практики

Скачать приложение

Тот же канал

Your AI Agent Starts Every Session Like a New Hire - Killian Carlsen-Phelan, Sonar

The Multi-Agent Architecture That Actually Ships — Luke Alvoeiro, Factory

Anthropic's CCA Exam as a Field-Guide for Agentic Engineering — Frank Coyle, UC Berkeley

Why Software Factories Fail

How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS

Beyond Code Coverage: Functionality Testing with Playwright — Marlene Mhangami, Microsoft

"Software Fundamentals Matter More Than Ever" — Matt Pocock

Don't Build Agents, Build Skills Instead – Barry Zhang & Mahesh Murag, Anthropic

Рекомендуемое видео

A Practical Guide To Becoming An AI Engineer (2026)

The Only SaaS Moats Left in 2026

Is AI Killing Human Fashion Design? Who Will Shape the Future of Design? Fashion Tea #6

How AI Empowers Tech Leaders Siwei Shen

Why Abu Dhabi Wants to Export Energy AI

Agentic AI Frameworks Explained: Workflows, Multi-Agent, & Production

Inside the secret race to build artificial intelligence | The Gray Area

It’s not just you

What AI Can Never Replace w/ Megan Kobzej | Cloud English Podcast Episode 91

When AIs act emotional

Harnessing generative AI in architecture, engineering & construction

Лексика и советы по произношению к этому уроку

Ключевая лексика этого видео

Фразовые глаголы, которые вы услышите

Грамматика в этом видео

Произношение, на которое стоит обратить внимание

Как заниматься с этим видео

Что такое техника Shadowing?