Pratik Goyal SUMMARY

A detail oriented and commercially aware professional with 4+ years of

experience and extensive knowledge in inspecting, cleaning,
EMAIL transforming, and modeling data with the goal of highlighting useful
[email protected] information.
Seeking a data scientist position to utilize my analytical skill and improve
business opportunities and growth.

Completed the Certificate Program in Big Data Analytics &

TELEPHONE Optimization at International School of Engineering (INSOFE),
7207897931 Hyderabad. The program is certified for the quality and assessment by
LTI of Carnegie Mellon University (CMU).
SKILLS

Database
TECHNICAL SKILLS
Text Mining
Machine Learning
Modelling Algorithms
Scripting Databases MySQL, MongoDB, Postgres
Statistical Analytics Machine Learning Classification, regression, clustering,
Communication Recommendation, anomaly detection,
feature engineering
Modelling Time Series Analysis, SVM, HMM,
Algorithms Neural Nets, Naive Bayes, K-nearest Neighbors,
Decision Trees, Random Forest, Genetic
Algorithm,
Montecarlo, TensorFlow
Scripting Python, R, R-shiny
Languages
Web Frameworks Flask, jQuery
Visualization tools GGplot, PowerBI

EXPERIENCE

Aptagrim Consulting LLP (September 2017 - Till Date)

Aptagrim specialize in helping organizations gain insights from their data and use those insights to change the way a
decision is made. We do this through a blend of machine learning, statistical analysis, data visualization, optimization, and
data engineering techniques.

1
 Pratik Goyal
Role : Data Scientist
• Hands- on experience in Build Machine Learning and deep Learning Models using Python, R.
• Experience in object-oriented programming (OOP) concepts using Python and R.
• Knowledge in SQL databases MySQL and Postgres.
• Experience in creating initial prototype Using Flask or Tkinter interface.
• Displayed and proved capability to learn any new technology and work on it.
• Ability to successfully multi-task and prioritize work.
• Knowledge in using Flask and Shiny web frame works.
• Hands on Experience on hosting the Django application on AWS server using Nginx and Gunicorn.

Project #1
Project Name: AI-powered smart assistive glasses for visually impaired
Tools Used: Python, Django, Html, CSS, Android, Deep learning (Kera’s tensorflow), OpenCV.

Project Description:
India being a capital to visually impaired people, challenges us to work on this problem by creating a product which can
help blind people to understand surroundings and identify people around them.

Our approach:
• We have leveraged deep learning technology and build glasses which will help a blind person to understand the
surroundings, also help them to identify the people, gender, and emotions.
• To build the machine learning and deep learning models which solves the problem statement, we have used
python and tensorflow to build models.
• We have integrated the tensorflow models in Android by converting the models from tensorflow to tensorflow
lite.

Project #2
Project Name: Applicant tracking System
Tools Used: Python, Django, Html, CSS, Pandas, Sklearn, SciPy, Tensorflow, spacy

Project Description:
1. The biggest issue for recruiters is that they must go through hundreds of resumes, which is laborious job, also it is
very difficult to identify the right skilled candidate for the organization. By the time recruiters selects the candidate
for an interview most of the highly skilled applicants have already taken up the new engagements.

To solve such problems, we have utilized machine to read numerous resumes to match the requirements.

2. In addition to the above, we have also observed that there is a high chance of receiving manipulated resumes due
to which the score of the resumes with respect to the job description is high/low therefore it becomes difficult to
predict whether the candidate is culturally fit or unfit to the organization.

To solve this problem we have built a psychometric model with the help of deep learning models.

2
 Pratik Goyal
Our approach:
• For a machine to understand the resume we have built a machine learning model which scans and analyze the
resume and provides the score for that applicant against the Job description.
• We have also built the user dashboard which helps to manage the jobs and CV’s and schedule interviews with
right candidate.
• It also gives the basic analytics graph on each candidate.
• Psychometric models take in the videos of candidate and analyze them and predict the Big five traits of the person.

Project #3
Project Name: Honey Encryption
Tools Used: Python, flask, Pandas, Sklearn, SciPy, Tensorflow.

Project Description:
Now-a-days we frequently use Instant Messaging (IM) system for communication. While it provides us convenience to
interact with others, there exist some side effects like an invasion of privacy through eavesdropping. Thus, popular IM
systems such as Telegram and have strengthened their security by message encryption. It makes the message unreadable
to anybody except the sender and the receiver. However, it is still weak in a brute-force attack because the security of
cryptosystem depends on the key size which has potential vulnerability to be cracked.

Our approach:
• We have built the encoder which understands the natural language sentence and replaces the message with
humanly understandable yet grammatically and syntactically simplified sentence having the same meaning for
humans which is decoded at the receiver end.
• We have used the NLP, Deep learning models to build the encoder.
• The basic flask interface was developed for a prototype.

Project #4
Project Name: Manuka Honey Forecast
Tools Used: Python, Django, Html, CSS, Pandas, Sklearn, SciPy.

Project Description:
Manuka honey is used as a natural ointment for wounds of all kinds. It has been hailed as a go-to germ fighter in an age
of resistance to conventional antibiotics. Proponents also claim that Manuka honey can treat other conditions from acne
to sinus issues.
Need to understand the expected yield from each field based on multiple parameters of weather, field type, past yield
and other conditions.

Our approach:
• We have built the tools which helps business to understand the expected yield from fields for Manuka honey, by
using the analytical graphs which also help them to make the decision by using machine learning model which
will predict the amount of manuka present in the honey.

3
 Pratik Goyal
• We also build model to predict the revenue from each field based on expected production and applying
mathematical calculations on top of this parameter.

Project #5
Project Name: Threat detection
Tools Used: R-Shiny Dashboard, HTML.

Project Description:
The insider threat is the biggest cyber security problem for organizations because insiders tend to cause the most damage.
They are also harder to detect and prevent in comparison to attacks from the outside. Insiders are just that – insiders with
keys to the kingdom. They know where the sensitive company/customer data is, and they have access to it. That means
insiders know exactly where to strike if they decide to act.
To solve this problem, we have built the shiny dashboard where a user can upload the daily log to see each employee’s
behavior.

Our approach:
• We have built statistical model which will detect the anomalies from the daily log file and the behavior of each
employee.

EDUCATION

Central Board of Secondary Education (54%) Kendriya Vidyalaya

2011

Board of Intermediate (67%) Ratna Jr. College

2013

Bachelor of Computer Application (70%) Osmania University, Bhavans Degree College

2016

Certificate Program in Big Data Analytics & Optimization, INSOFE

2017

4
 Pratik Goyal

PERSONAL INFORMATION

Date of Birth 26th September 1995

Marital Status Married
Languages Known Hindi, English,
Father’s Name Vimal Kumar