ABOUT THE SPEAKER
Priyanka Jain - Technologist
Priyanka Jain heads up product for pymetrics, an NYC-based startup that uses neuroscience and AI to make hiring more diverse and effective.

Why you should listen

Passionate about using technology to create a fairer workplace and global economy, Priyanka Jain is a spokesperson for the United Nations Foundation's Girl Up Campaign, Chair of the Acumen Fund's Junior Council and on the Innovation Board for the XPrize Foundation. She received her B.S. from Stanford University, where she was President of Stanford Women in Business and one of 12 Mayfield Entrepreneurship Fellows. Her previous experience includes internships at IBM Watson, Shift Technologies, Canvas Ventures and the Institute for Learning and Brain Sciences. Outside of work, she loves playing tennis and eating anything covered in dark chocolate.

More profile about the speaker
Priyanka Jain | Speaker | TED.com
The Way We Work

Priyanka Jain: How to make applying for jobs less painful

朴雅卡· 耆那: 怎样让工作申请少一些痛苦

Filmed:
548,985 views

找工作时通常先要把你的简历提交给数不清的雇主,然后大多数永远不会有回应。但是越来越多的公司正在使用前沿的技术识别候选人。如果AI成为招聘的未来,它对你将意味着什么?技术专家朴雅卡· 耆那向我们展示了这种新的招聘场景。
- Technologist
Priyanka Jain heads up product for pymetrics, an NYC-based startup that uses neuroscience and AI to make hiring more diverse and effective. Full bio

Double-click the English transcript below to play the video.

00:00
Applying应用 for jobs工作 online线上
0
548
1416
在线申请工作
是我们这个时代最糟糕的
数字化体验之一。
00:01
is one of the worst最差
digital数字 experiences经验 of our time.
1
1988
2616
面对面交谈也没好到哪儿去。
00:04
And applying应用 for jobs工作 in person
really isn't much better.
2
4628
2696
【我们的工作方式】
00:07
[The Way We Work]
3
7349
1720
00:11
Hiring招聘 as we know it
is broken破碎 on many许多 fronts战线.
4
11588
2456
众所周知,招聘方式
在很多方面一团糟。
00:14
It's a terrible可怕 experience经验 for people.
5
14068
1856
对人们来说这是一个糟糕的经历。
在过去一年
00:15
About 75 percent百分 of people
6
15948
1816
使用多种方式申请工作时的
群体中,大约有75%的人
00:17
who applied应用的 to jobs工作
using运用 various各个 methods方法 in the past过去 year
7
17788
2856
说他们从未收到雇主的任何反馈。
00:20
said they never heard听说 anything back
from the employer雇主.
8
20668
2656
对公司来说,这不是一件好事情。
00:23
And at the company公司 level水平
it's not much better.
9
23348
2176
在开始工作的不到一年时间里,
00:25
46 percent百分 of people get fired解雇 or quit放弃
10
25548
2776
46%的人被解雇或者主动离职。
00:28
within the first year
of starting开始 their jobs工作.
11
28348
2176
这一点很令人震惊。
00:30
It's pretty漂亮 mind-blowing令人兴奋.
12
30548
1216
这种现象对经济也产生了负面影响。
00:31
It's also bad for the economy经济.
13
31788
1456
在历史上第一次,
00:33
For the first time in history历史,
14
33268
1456
招聘岗位超过了无业人员的人数,
00:34
we have more open打开 jobs工作
than we have unemployed失业的 people,
15
34748
2856
对我而言,这意味着出问题了。
00:37
and to me that screams尖叫声
that we have a problem问题.
16
37628
2176
我认为这一切的关键在于一张纸:
简历。
00:39
I believe that at the crux症结 of all of this
is a single piece of paper: the résumé.
17
39828
3976
毫无疑问,简历中包含着一些
有用的信息:
00:43
A résumé definitely无疑 has
some useful有用 pieces in it:
18
43828
2336
人们扮演过哪些角色,
有哪些计算机技能,
00:46
what roles角色 people have had,
computer电脑 skills技能,
19
46188
2096
精通什么语言,
00:48
what languages语言 they speak说话,
20
48308
1256
但并未提及他们有哪方面的潜力,
00:49
but what it misses错过 is
what they have the potential潜在 to do
21
49588
3056
这些事情他们在过去可能没机会去做。
00:52
that they might威力 not have had
the opportunity机会 to do in the past过去.
22
52668
2976
在变化如此迅速的经济环境中,
在线发布的工作机会
00:55
And with such这样 a quickly很快 changing改变 economy经济
where jobs工作 are coming未来 online线上
23
55668
3256
可能要求的都是没人掌握的技术,
00:58
that might威力 require要求 skills技能 that nobody没有人 has,
24
58948
2056
如果我们只看一个人过去做了什么,
01:01
if we only look at what someone有人
has doneDONE in the past过去,
25
61028
2776
就不能把这个人和
未来的工作匹配起来。
01:03
we're not going to be able能够
to match比赛 people to the jobs工作 of the future未来.
26
63828
3256
所以我认为这是技术真正有用的地方。
01:07
So this is where I think technology技术
can be really helpful有帮助.
27
67108
2736
您可能已经看到了算法如何很好的
01:09
You've probably大概 seen看到
that algorithms算法 have gotten得到 pretty漂亮 good
28
69868
2856
把人和事物匹配起来,
01:12
at matching匹配 people to things,
29
72748
1536
但是如果我们把同样的技术用于
01:14
but what if we could use
that same相同 technology技术
30
74308
2256
真正帮助找到那些为
我们量身打造的工作昵?
01:16
to actually其实 help us find jobs工作
that we're really well-suited非常适合 for?
31
76588
3096
我知道你在想什么。
01:19
But I know what you're thinking思维.
32
79708
1576
让算法为你挑拣下一份工作
听起来有点离谱,
01:21
Algorithms算法 picking选择 your next下一个 job工作
sounds声音 a little bit scary害怕,
33
81308
2776
但有个东西已经被证明
01:24
but there is one thing that has been shown显示
34
84108
2056
能够成功预测某人
是否能胜任未来的工作,
01:26
to be really predictive预测
of someone's谁家 future未来 success成功 in a job工作,
35
86188
2896
这就是所谓的多评估测试。
01:29
and that's what's called
a multimeasure多措施 test测试.
36
89108
2136
多评估测试并不是什么新概念,
01:31
Multimeasure多措施 tests测试
really aren't anything new,
37
91268
2176
但是它们曾经价格不菲,
01:33
but they used to be really expensive昂贵
38
93468
1736
并且需要一个博士坐在你对面,
01:35
and required需要 a PhD博士 sitting坐在 across横过 from you
39
95228
2016
回答一堆问题并且整理成报告。
01:37
and answering回答 lots of questions问题
and writing写作 reports报告.
40
97268
2456
多评估测试是一种用来
01:39
Multimeasure多措施 tests测试 are a way
41
99748
1696
理解某人内在特质的方法——
01:41
to understand理解 someone's谁家 inherent固有 traits性状 --
42
101468
2456
你的记忆力,你的专注力。
01:43
your memory记忆, your attentiveness注意力.
43
103948
1776
01:46
What if we could take multimeasure多措施 tests测试
44
106662
1942
如果我们能够做多评估测试,
让公众都可以参与,
01:48
and make them scalable可扩展性 and accessible无障碍,
45
108628
2536
并且把相关数据提供给雇主,
比如某个人的某些特质
01:51
and provide提供 data数据 to employers雇主
about really what the traits性状 are
46
111188
3376
使其真的很适合这个工作,会怎样?
01:54
of someone有人 who can make
them a good fit适合 for a job工作?
47
114588
2896
这些听起来很抽象。
01:57
This all sounds声音 abstract抽象.
48
117508
1296
让我们一起试试其中一个游戏。
01:58
Let's try one of the games游戏 together一起.
49
118828
1735
你将要看到一个闪烁的圆,
02:00
You're about to see a flashing闪烁 circle,
50
120587
1857
你的任务就是当圆是红色时鼓掌,
02:02
and your job工作 is going to be
to clap when the circle is red
51
122468
2960
02:06
and do nothing when it's green绿色.
52
126285
1496
当圆是绿色时什么也不做。
【准备好了?】
02:07
[Ready准备?]
53
127806
1376
【开始】
02:09
[Begin开始!]
54
129207
1360
02:11
[Green绿色 circle]
55
131708
1000
【绿色圆】
02:13
[Green绿色 circle]
56
133708
1040
【绿色圆】
02:15
[Red circle]
57
135708
1000
【红色圆】
02:17
[Green绿色 circle]
58
137708
1080
【绿色圆】
02:19
[Red circle]
59
139708
1000
【红色圆】
02:21
Maybe you're the type类型 of person
60
141928
1596
或许你是那种
在红色圆出现后毫秒内鼓掌的人。
02:23
who claps拍手 the millisecond毫秒
after a red circle appears出现.
61
143548
2496
或者你是另外一种人,
02:26
Or maybe you're the type类型 of person
62
146068
1656
那种需要多花点时间,
等到100%确认才行动的人。
02:27
who takes just a little bit longer
to be 100 percent百分 sure.
63
147748
2735
或者你在还不确定时
就为绿色圆鼓掌。
02:30
Or maybe you clap on green绿色
even though虽然 you're not supposed应该 to.
64
150508
2936
很酷的一点是这并不
像是个标准的测试,
02:33
The cool thing here is that
this isn't like a standardized标准化 test测试
65
153468
2976
那种决定能被雇佣与否的测试。
02:36
where some people are employable受 雇
and some people aren't.
66
156468
2656
相反,这是个理解你的特性和
02:39
Instead代替 it's about understanding理解
the fit适合 between之间 your characteristics特点
67
159148
3256
适合你的工作之间的匹配度测试。
02:42
and what would make you
good a certain某些 job工作.
68
162428
2016
我们发现如果你在红色时鼓掌晚,
而在绿色时从不鼓掌,
02:44
We found发现 that if you clap late晚了 on red
and you never clap on the green绿色,
69
164468
3736
你可能具备高度专注力,
能够很好的自我约束。
02:48
you might威力 be high in attentiveness注意力
and high in restraint克制.
70
168228
3176
在那个象限的人们
往往擅长学习和考试,
02:51
People in that quadrant象限 tend趋向 to be
great students学生们, great test-takers考生,
71
171428
3576
精于项目管理和财会。
02:55
great at project项目 management管理 or accounting会计.
72
175028
2136
如果你在红色时立刻鼓掌,
并且有时在绿色鼓掌,
02:57
But if you clap immediately立即 on red
and sometimes有时 clap on green绿色,
73
177188
3336
那意味着你可能易冲动
并且具备创造性,
03:00
that might威力 mean that
you're more impulsive浮躁 and creative创作的,
74
180548
2656
我们发现顶级的商人
经常会表现出这些特质。
03:03
and we've我们已经 found发现 that top-performing表现最出色
salespeople销售人员 often经常 embody体现 these traits性状.
75
183228
3875
我们在招聘中使用它的方式是
03:07
The way we actually其实 use this in hiring招聘
76
187128
2016
我们让角色中表现出色的人参与
03:09
is we have top最佳 performers表演者 in a role角色
go through通过 neuroscience神经科学 exercises演习
77
189168
3696
类似的神经科学训练。
03:12
like this one.
78
192888
1216
然后我们开发了一个算法
03:14
Then we develop发展 an algorithm算法
79
194128
1376
来理解是什么让这些
表现出众者脱颖而出。
03:15
that understands理解 what makes品牌
those top最佳 performers表演者 unique独特.
80
195528
2656
然后当人们申请工作的时候,
03:18
And then when people apply应用 to the job工作,
81
198208
1936
我们就会优先列出
最适合那项工作的候选人。
03:20
we're able能够 to surface表面 the candidates候选人
who might威力 be best最好 suited合适的 for that job工作.
82
200168
4136
你可能在思考其中存在的风险。
03:24
So you might威力 be thinking思维
there's a danger危险 in this.
83
204328
2376
当今的职场多样性仍有待提高,
03:26
The work world世界 today今天
is not the most diverse多种
84
206728
2136
如果我们基于当下的
出众员工构建算法,
03:28
and if we're building建造 algorithms算法
based基于 on current当前 top最佳 performers表演者,
85
208888
3096
要怎样确保
03:32
how do we make sure
86
212008
1216
我们不是在固守既有的偏见呢?
03:33
that we're not just perpetuating延续
the biases偏见 that already已经 exist存在?
87
213248
2976
例如,如果我们基于顶尖表现的
CEO构建一个算法
03:36
For example, if we were building建造
an algorithm算法 based基于 on top最佳 performing执行 CEOs老总
88
216248
4056
并且使用S&P500作为一个训练集,
03:40
and use the S&amp功放;P 500 as a training训练 set,
89
220328
3216
你将会发现
03:43
you would actually其实 find
90
223568
1256
更可能雇佣一个叫约翰的
白人男子而非任何女性。
03:44
that you're more likely容易 to hire聘请
a white白色 man named命名 John约翰 than any woman女人.
91
224848
3816
这是目前谁正处在这个角色的现实。
03:48
And that's the reality现实
of who's谁是 in those roles角色 right now.
92
228688
2696
但是技术实际上给出了
一个真正有趣的机会。
03:51
But technology技术 actually其实 poses姿势
a really interesting有趣 opportunity机会.
93
231408
3376
我们可以创造一些比人类
任何时候都更平等
03:54
We can create创建 algorithms算法
that are more equitable公平
94
234808
2256
和更公正的算法。
03:57
and more fair公平 than human人的 beings众生
have ever been.
95
237088
2256
每一个我们投入生产的
算法都会被预先进行测试
03:59
Every一切 algorithm算法 that we put
into production生产 has been pretested预先测试
96
239368
3696
以确保它不会偏爱任何性别
或者种族。
04:03
to ensure确保 that it doesn't favor偏爱
any gender性别 or ethnicity种族.
97
243088
3096
如果有任何人群正在被过度偏爱,
04:06
And if there's any population人口
that's being存在 overfavored过度青睐,
98
246208
2736
我们可以调整算法直到该现象消失。
04:08
we can actually其实 alter改变 the algorithm算法
until直到 that's no longer true真正.
99
248968
3120
04:12
When we focus焦点 on the inherent固有
characteristics特点
100
252448
2216
当我们关注在那些让一个人
非常适合一个工作的内在特质时,
04:14
that can make somebody
a good fit适合 for a job工作,
101
254688
2096
我们可以超越种族,阶级,
性别和老龄化主义——
04:16
we can transcend超越 racism种族主义,
classismclassism, sexism性别歧视, ageism年龄歧视 --
102
256808
3576
甚至是名校背景。
04:20
even good schoolism学校主义.
103
260408
1416
我们最好的技术和算法不应该只用于
04:21
Our best最好 technology技术 and algorithms算法
shouldn't不能 just be used
104
261848
2896
帮助寻找我们的下一个卖座电影
或者贾斯汀·比伯的新歌。
04:24
for helping帮助 us find our next下一个 movie电影 binge狂欢
or new favorite喜爱 Justin贾斯汀 Bieber比伯 song歌曲.
105
264768
3736
想象一下如果我们能够利用技术的
力量,
04:28
Imagine想像 if we could harness马具
the power功率 of technology技术
106
268528
2656
在更深层次上理解我们是谁,
并得到一个
04:31
to get real真实 guidance指导
on what we should be doing
107
271208
2296
我们应该做什么的真正指引会怎样。
04:33
based基于 on who we are at a deeper更深 level水平.
108
273528
1936
Translated by Jingle duan
Reviewed by jacks jun

▲Back to top

ABOUT THE SPEAKER
Priyanka Jain - Technologist
Priyanka Jain heads up product for pymetrics, an NYC-based startup that uses neuroscience and AI to make hiring more diverse and effective.

Why you should listen

Passionate about using technology to create a fairer workplace and global economy, Priyanka Jain is a spokesperson for the United Nations Foundation's Girl Up Campaign, Chair of the Acumen Fund's Junior Council and on the Innovation Board for the XPrize Foundation. She received her B.S. from Stanford University, where she was President of Stanford Women in Business and one of 12 Mayfield Entrepreneurship Fellows. Her previous experience includes internships at IBM Watson, Shift Technologies, Canvas Ventures and the Institute for Learning and Brain Sciences. Outside of work, she loves playing tennis and eating anything covered in dark chocolate.

More profile about the speaker
Priyanka Jain | Speaker | TED.com

Data provided by TED.

This site was created in May 2015 and the last update was on January 12, 2020. It will no longer be updated.

We are currently creating a new site called "eng.lish.video" and would be grateful if you could access it.

If you have any questions or suggestions, please feel free to write comments in your language on the contact form.

Privacy Policy

Developer's Blog

Buy Me A Coffee