1 of 50

Why are you here

today?

2 of 50

by

Sumit Tyagi

How to build successful AI career?

3 of 50

Agenda

AI landscape

​

Roles in current AI landscape

​

How to decide your role?

​

What would you need to achieve it?

​

​

​

​

4 of 50

🤖

Lead Data Scientist @ Paytm

👨‍🏫

Visiting AI Faculty @ INSAID, UpGrad, KIMO, etc.

🏋️

Can bench-press 100 KGs

About me

Consulting @ Absolutdata - Infogain

Product @ LEAD School

💼

💻

🤓

SME @ UpGrad

Previously…….

5 of 50

AI Landscape

What do you think humans want ?

6 of 50

AI Landscape

Humans wants to relax.

7 of 50

AI Landscape

Humans wants to relax,

But they have work to do.

8 of 50

AI Landscape

Humans wants to relax, But they have work to do.

So they either delegate or automate the work.

9 of 50

AI Landscape

Remember,

​

  • If you can write the solution of a task on a paper, it can be automated.
  • It’s called rule-based approach.

​

  • If it’s difficult for you to write it down.
  • You may need to delegate it to a person or AI.

​

10 of 50

Quickdraw Game

g.co/quickdraw

g.co/� quick� draw

11 of 50

Quickdraw Game - Discussion

  1. How does the game work?
  2. How is it recognising your drawings?
  3. Further enquiry: How could we program this?

12 of 50

How does ML work in QuickDraw?

g.co/quickdrawdata

g.co/� quick� draw� data

13 of 50

Quickdraw: Rule-based

​

if object.height > 10:

do x

if object.color is blue:

do y

if object.numberOfLegs > 2:

do z

...

Frame the goal of the product

Refine until goals are met

Designers and engineers develop flow and logic

14 of 50

Quickdraw: Machine Learning

Frame the goal of the product

Refine until goals are met

Train a model using examples

15 of 50

AI Landscape

16 of 50

Quickdraw Game

g.co/quickdraw

g.co/� quick� draw

17 of 50

Which Approach to Use?

Rules-based approach

Machine Learning

Alphabetising a list of song titles

Rules-based

18 of 50

Which Approach to Use?

Rules-based approach

Machine Learning

Ranking Web Search Results

Rules-based

Machine Learning

19 of 50

Which Approach to Use?

Rules-based

Machine Learning

Predicting Housing Prices�based on Location

Machine Learning

20 of 50

Which Approach to Use?

Processing online payments

Rules-based approach

Machine Learning

Rules-based

21 of 50

Which Approach to Use?

Classifying an object in a photo

Rules-based

Machine Learning

Machine Learning

22 of 50

Recap

Rule-based Approach

Machine Learning

  • Rules are defined
  • Improvements come from algorithms and network
  • Learns patterns from data
  • Improvements can from additional data

Each has its benefits

AI Landscape

23 of 50

Roles in AI landscape

24 of 50

Roles in AI Landscape

Maths

25 of 50

Roles in AI Landscape

OCD

Maths

26 of 50

Roles in AI Landscape

OCD

Maths

Dressing sense

27 of 50

Roles in AI Landscape

OCD

Maths

Dressing sense

AI Researcher

28 of 50

Roles in AI Landscape

Paint

29 of 50

Roles in AI Landscape

Design

Paint

30 of 50

Roles in AI Landscape

Design

Paint

Aesthetics

31 of 50

Roles in AI Landscape

Design

Paint

Aesthetics

Maths

32 of 50

Roles in AI Landscape

Design

Paint

Aesthetics

Data Visualizer

Maths

33 of 50

Roles in AI Landscape

Maths

34 of 50

Roles in AI Landscape

Ctrl C

+

Ctrl V

Maths

35 of 50

Roles in AI Landscape

Ctrl C

+

Ctrl V

Code

Maths

36 of 50

Roles in AI Landscape

Ctrl C

+

Ctrl V

Code

Maths

Blame

Game

37 of 50

Roles in AI Landscape

Ctrl C

+

Ctrl V

Code

Data Scientist

Maths

Blame

Game

38 of 50

Roles in AI Landscape

Code

39 of 50

Roles in AI Landscape

Cloud

Code

40 of 50

Roles in AI Landscape

Cloud

I can fix it later

Code

41 of 50

Roles in AI Landscape

Cloud

I can fix it later

Code

Data

42 of 50

Roles in AI Landscape

Cloud

I can fix it later

Data Engineer

Code

Data

43 of 50

Roles in AI Landscape

Cloud

It works on my machine

ML Engineer

Code

Data

44 of 50

Roles in AI Landscape

Cloud

It works on my machine

MLOps Engineer

Code

Data

DevOps

45 of 50

What to avoid?

46 of 50

What to avoid?

  • Decide your mentors wisely.

​

​

47 of 50

What to avoid?

  • Decide your mentors wisely.

​

  • Logic is important.

48 of 50

What to avoid?

  • Decide your mentors wisely.

​

  • Logic is important.

​

  • Referrals are important.

​

​

49 of 50

What to avoid?

  • Decide your mentors wisely.

​

  • Logic is important.

​

  • Referrals are important.

​

  • Give your 100% everyday.

50 of 50

  • Let’s catchup after the session.
  • Click a lot of photos.
  • Ask for referral.

LinkedIn

Twitter