Aninda Kumar Sharma

Phone: 01760755706
E-mail: [email protected]
Hyperion Green Park 1, Rd No. 2, Mirpur 11.5, Dhaka 1216

Career Objectives

I am a computer systems engineer and a passionate machine learning enthusiast with interest in
computer engineering, predictive modeling and multimedia systems. Within the span of two years, I
have worked in the software industry with a focus on machine learning and computer systems. Moving
forward, I plan to join academics to devote myself completely to research and teaching. My
competencies include developing machine learning algorithms, data analysis, predictive modeling and
language-based problems development that I implemented in my previous jobs. I have also been a part
of several research projects, with some papers already undergoing peer review in top journals. To this
end, I believe I would be able to pursue MSc and PhD education in the fields of my interest which
include artificial intelligence and multimedia systems. In the end, my goal is to carry out innovative
research and make sure that there is a solid connection between theory and its use, particularly in the
case of AI and machine learning in multimedia systems.

Portfolio
SL Website Project Link

1 Github https://github.com/Aninda11?tab=repositories

2 Linkedin https://www.linkedin.com/in/aninda-kumar-sharma-a81292235/

3 Research Gate https://www.researchgate.net/profile/Aninda-Sharma

Work Experience
Machine Learning Intern (08 September 2024 – Present) —Acme AI, Dhaka, Bangladesh

● HR Chatbot Development: I’m designing and implementing a Retrieval-Augmented Generation

(RAG) based chatbot for Acme Technologies, focusing on integrating HR-specific data for efficient
query handling.

● Site Monitoring Project: I’m developing a face recognition system for a site monitoring project,
incorporating models for detecting violations and recognizing individuals from video streams,
specifically using YOLO and FGG-16.

● System Integration and Validation: I’m ensuring smooth integration of various components, including
safety checks and edge detection, and implementing validation systems for real-time monitoring and
reporting of detected events.

● User Testing and Deployment: I am conducting thorough testing and gathering feedback to refine the
chatbot and site monitoring systems before their official launch, ensuring they meet user needs and
performance standards.

Associate Software Engineer, Python (08 December 2022 – 31 August 2024) — Mevrik DCX,
Dhaka, Bangladesh

● I have been deeply engaged in the development of an AI-powered All-in-one platform to connect,
engage customers in digital and social channels. My focus has been on creating interactive chat bots that
can engage users in a dynamic and responsive manner. I have also worked on integrating various APIs
into these applications to enhance their functionality and provide a more comprehensive user experience.

● I also researched Data Science to enhance our product and the company’s overall growth.

● My responsibilities extend across the entire product development lifecycle. This includes participating
in the initial design phase, where I contribute to shaping the product's features and overall structure. I
then actively develop these features, ensuring they are implemented effectively and efficiently.

● I have also had the opportunity to work on integrating Artificial Intelligence into our platform. This
involves training data for chat bots, which helps improve their accuracy and efficiency. By incorporating
AI, we are able to create more intelligent and responsive chat bots that can better serve our users.

Remote Data Science Intern (5 March 2024 – 31 June 2024) — iNeuron.ai, Bengaleru, India

● Development of a model based on Dynamic Convolutional Neural Network (DCNN) to efficiently

detect malicious URLs. This model was designed to identify and defend against threats such as phishing
and spam attacks.

● Conduction of a case study on the use of Transformer models in cybersecurity. The focus was on
detecting malicious domain names. This study contributed to improving the efficiency and accuracy of
threat detection.

● Implementation of a system that automatically sends alerts to the cybersecurity team as soon as a
malicious URL is detected. Additionally, I maintained a database to store all detected malicious URLs.
This database serves as a valuable resource for future reference and analysis.

Virtual Cyber Security Intern (5 September 2022 – 5 December 2022) — Creative IT Institute,
Dhaka

● I have been working here to develop my knowledge on Cyber Security under a supervisor. ● Total 6
project works were undertaken which included works on web application vulnerability testing, system
vulnerability testing and information gathering.

● Learnt about writing Penetration Test Reports for 3 vulnerable operating systems and 3 web
applications.
● I worked to develop content for the Cyber Security and General Knowledge section for the ICT
Olympiad Bangladesh organization. We designed the syllabus from which students would study about
the most fascinating areas of cyber security

Education Certification
Exam Title Subject/ Institute/University/Boar Result Passi Duration Date
Group d ng
Year

BSc Computer American International 3.44 2022 4 years January 14,

Engineering University of 2022
Bangladesh

HSC Science Dhaka Board 4.17 2017 2 years July 29,

2017

SSC Science Dhaka Board 4.78 2015 2 years May 30,

2015

Project Work (s)

IoT Based Smart Flood Monitoring System (BSc Capstone Project, SEMESTER: Fall 21-22): We
designed a system that can detect humidity, temperature, rain, water level, and water growth rate in
order to create a smooth flood detection system. A machine learning model was designed that predicts
the occurrence of flood occurrences based on previously gathered data in the project.

Small Various Object Detection using YOLO5 (BSc COURSE – CVPR, SEMESTER: Fall
21-22): In this paper, we have shown that the task of detecting small objects can be made easier, at
least in natural scenarios, if we leverage the specific circumstances of this domain: Since the context is
usually clearly defined and well-structured, we can make use of it in the object detection process by
first recognizing the context of a small object and then the object itself.
Floor Cleaning Robot (BSc COURSE – Microprocessor and Embedded System , SEMESTER:
Fall 21- 22) : We made a floor-cleaning robot in our project. The floor-cleaning robot could
automatically clean the floor. We used a cloth pad and a water supply system. The robot needs to cover
the whole floor to clean, so the robot has a wheel to move around and an algorithm that covers the
whole surface of the floor.
Intelligent System for an insurance provider.(PostGrad COURSE – Data Science, SEMESTER:
Summer 23-24) This project's primary objective is to develop an intelligent system for an insurance
provider. Our goal is to determine the probability that clients would commit fraud or other covert acts
using their insurance. We'll also look at the potential costs to the business in the event that our prediction
system is off. The report, directly integrated into the Jupyter notebook, will provide a clear explanation
of everything and demonstrate the functionality of our system.
 Research Work(s)

I have authored and co-authored 4 research papers, 2 are currently on peer review stage in Q1 journals
and among these, 1 has been approved as a preprint by Springer Nature as of October 2024. I also wrote
an official research project book during my undergraduate capstone project.

1. “A Systematic Literature Review on Bengali Grammar Correction System” (Preprint - Springer

Language Resources and Evaluation - September 2024) :
This paper presents a comprehensive review of various methods and systems developed for Bengali
grammar correction. It surveys different machine learning, rule-based, and hybrid approaches to correct
grammar issues specific to the Bengali language.

2. "Smart Early Flood Monitoring System Using IoT" (Faculty of Engineering, AIUB - January 2022)
This project proposes an intelligent flood monitoring system that integrates IoT-enabled sensor networks
and machine-learning models to forecast floods.It uses environmental data such as humidity,
temperature, water level, and rainfall to detect floods. Flood prediction model was based on MATLAB.

3. Automatic Bangladeshi Number Plate Recognition Using Multimodal Deep Learning (Accepted,
2024 27th IEEE International Conference on Computer and Information Technology)
We applied a deep neural network model to the input image to identify the region of interest. The
segmentation of the characters on license plates likewise uses the same kind of methodology. Then, to
recognize the characters of the license plates a different neural network is used. The construction of a
system using multiple neural networks to recognize number plates that are written in Bengali is what
makes this research unique.

4. DistilBERT Based Malicious URL Detection System (Accepted, 2024 4th IEEE International
Conference on Digital Society and Intelligent Systems )
This study proposes a novel approach to identify malicious URLs using the DistilBERT transformer
model focusing on the improvement of detection accuracy. We built a complete data preprocessing
pipeline which preprocessed the URL features utilizing big data technology to analyze URL patterns and
anomalies. By doing so, the proposed system tends to categorize URLs into benign, defacement,
phishing or malware groups thus enhancing detection accuracy significantly.

Extra Curricular Technical Courses

Organization Course Title Certificate Date

 University of Programming for Link June 9, 2020
Michigan Everybody (Getting
(Coursera) Started with Python)

University of Python Data Link June 22, 2020

Michigan Structures
(Coursera)

University of Machine Learning Link November 15, 2022

Washington Foundations: A Case
(Coursera) Study Approach

Creative IT Institute Ethical Hacking Link November 15, 2022

Hard Skills

Tool Knowledge: Jupyter Notebook, Pycharm, VMware Fusion, Jupyter Notebook, Google Colab,
Matlab, Simulink

Industry Knowledge: Machine Learning, Data Mining, Predictive Analytics, Data Analysis, Cyber
Security, Google Cloud, Matlab

Technical Skills: Python, SQL, Django Application Framework(Python), Flask Application

Framework(Python), FastAPI(Python), Pandas, Sklearn, Pytorch, Tensorflow, MATLAB, Kali Linux
Tools

Soft Skills
● Time Management
● Adaptability
● Creativity
● Hard working and goal oriented
● Communication

Personal Details:

• Father’s Name: Dr. Arun Kumar Sharma

• Mother’s Name Padma Sharma

 • Date of Birth 7 November 1998
• Gender Male

• Marital Status Unmarried

• Nationality Bangladesh

• Religion Hinduism

• Blood Group O+

• Permanent Address: Hyperion Green Park 1, Apartment 202, Plot # 1, 1A, Block # 2C, Rd No. 2,
Mirpur 11.5, Dhaka 1216
• Present Address Hyperion Green Park 1, Apartment 202, Plot # 1, 1A, Block # 2C, Rd No. 2,
Mirpur 11.5, Dhaka 1216

Reference (2 Persons)

#1

Name: Gadiel Myron

Company: Mevrik DCX

Designation: Product Head

Email address: [email protected]

Phone: 01645962192

#2
Name: MD. Saniat Rahman Zishan
Institution: American International University Of Bangladesh
Designation: Associate Professor, Department Head , Computer Engineering
Email address: [email protected]

Declaration

I sincerely declare that all the details shared above are true to my belief. I am answerable for its
correctness.
Aninda Kumar Sharma