Freelancer logo
Log In Sign Up
Post a Project

EXPLORE

AI Chatbot Development Artificial Intelligence Data Mining Python Web Scraping
Profile cover photo
User Avatar

Ahmed A.

@nexvative

monthly-level-three.svg
5.0
7

4.1

4.1

100%

Generative AI | Computer Vision | NLP

$20 USD / Hour

・

Flag of

Pakistan (6:44 PM)

・

Joined on October 19, 2024

$20 USD / Hour

・

Looking for a software development partner skilled in cutting-edge technologies like computer vision, generative AI, NLP, and machine learning? We've got you covered. Our expertise spans these domains and beyond, ensuring innovative, high-quality solutions tailored to your needs. With a commitment to staying ahead of industry trends, we deliver state-of-the-art results that drive success. Let us take your project to the next level with excellence, reliability, and a vision for the future.

Verifications

On time

100%

On budget

100%

Accept rate

100%

Repeat hire rate

10%

Freelancer Web Scraping Specialists Pakistan
See all Web Scraping Specialists in Pakistan

Portfolio

Portfolio

13035417
This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

13035399
VAE for Alzheimer’s MRI Augmentation This project focuses on advanced data augmentation for medical imaging, specifically targeting MRI brain scans used in the diagnosis of Alzheimer's disease. I implemented a Variational Autoencoder (VAE) using TensorFlow to learn the underlying distribution of the images class from a real-world Alzheimer’s dataset. The goal was to generate realistic, high-quality synthetic MRI images to enrich the training data and address class imbalance — a common challenge in medical datasets. The augmented data was used to improve the performance and generalization of deep learning models in Alzheimer's classification tasks. This technique is especially valuable in healthcare AI where data scarcity, privacy concerns, and class imbalance are major obstacles. This project showcases my skills in medical imaging, unsupervised learning, data augmentation, and deep learning workflows tailored for healthcare AI solutions.

VAE for Alzheimer’s MRI Augmentation

12344286
I developed an AI-driven Telegram bot that provides personalized career advice and mock interview sessions. Built using Google Gemini model and LangChain, this bot delivers tailored, real-time guidance for users across various industries and roles. Key Features: Real-Time Career Guidance: Uses advanced language models to give personalized career advice, adapting to individual goals and career paths. Mock Interview Practice: Offers interactive mock interviews with real-time feedback to help users refine responses and boost interview confidence. Personalized Memory: Integrated LangChain’s memory capabilities, allowing the bot to recall past interactions, providing consistent, context-aware advice. Optimized Conversational Flow: Applied advanced prompt engineering for natural, engaging conversations that align with user needs.

LLM Powered Telegram Chatbot

12305695
In this project, I implemented a U-Net architecture to accurately perform brain tumor segmentation from MRI scans. The primary goal was to isolate and highlight regions of tumor tissue, aiding in diagnosis and treatment planning. U-Net, a convolutional neural network (CNN) designed specifically for biomedical image segmentation, is highly effective in tasks where precise localization is required. The model was trained and validated on a dataset of brain MRI images, achieving high accuracy in segmenting tumor regions. The use of U-Net’s encoder-decoder architecture enabled efficient feature extraction and detailed segmentation, even with a limited amount of data. The results demonstrate the potential of deep learning in medical imaging, particularly in automating complex tasks like tumor detection. For more details and to explore the code, check out the project on my Kaggle notebook.

Brain Tumour Segmentation Using UNet.

In this project, I implemented a U-Net architecture to accurately perform brain tumor segmentation from MRI scans. The primary goal was to isolate and highlight regions of tumor tissue, aiding in diagnosis and treatment planning. U-Net, a convolutional neural network (CNN) designed specifically for biomedical image segmentation, is highly effective in tasks where precise localization is required. The model was trained and validated on a dataset of brain MRI images, achieving high accuracy in segmenting tumor regions. The use of U-Net’s encoder-decoder architecture enabled efficient feature extraction and detailed segmentation, even with a limited amount of data. The results demonstrate the potential of deep learning in medical imaging, particularly in automating complex tasks like tumor detection. For more details and to explore the code, check out the project on my Kaggle notebook.

Brain Tumour Segmentation Using UNet.

13912985
This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

13035417
This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

This project presents a custom-trained YOLO (You Only Look Once) model specifically engineered for robust face detection under highly variable conditions. The model is optimized to accurately detect human faces regardless of orientation (frontal, profile, upward, downward), image resolution, lighting conditions, or camera quality—a challenge that often limits off-the-shelf detectors. Key Features: Orientation-Invariance: Trained on a diverse dataset with extreme pose variations to ensure reliable face detection from any angle. Quality-Robustness: Performs consistently well even on low-resolution or noisy images, enabling use in surveillance footage or low-bandwidth video streams. Minimal Bias & Variance: Carefully curated and augmented training data led to a model with strong generalizability and minimal overfitting. Evaluation on external test sets showed excellent performance across age, gender, and ethnicity. Real-Time Capability: Leveraging the speed advanta

Robust Face Detection with YOLO

13035399
VAE for Alzheimer’s MRI Augmentation This project focuses on advanced data augmentation for medical imaging, specifically targeting MRI brain scans used in the diagnosis of Alzheimer's disease. I implemented a Variational Autoencoder (VAE) using TensorFlow to learn the underlying distribution of the images class from a real-world Alzheimer’s dataset. The goal was to generate realistic, high-quality synthetic MRI images to enrich the training data and address class imbalance — a common challenge in medical datasets. The augmented data was used to improve the performance and generalization of deep learning models in Alzheimer's classification tasks. This technique is especially valuable in healthcare AI where data scarcity, privacy concerns, and class imbalance are major obstacles. This project showcases my skills in medical imaging, unsupervised learning, data augmentation, and deep learning workflows tailored for healthcare AI solutions.

VAE for Alzheimer’s MRI Augmentation

12344286
I developed an AI-driven Telegram bot that provides personalized career advice and mock interview sessions. Built using Google Gemini model and LangChain, this bot delivers tailored, real-time guidance for users across various industries and roles. Key Features: Real-Time Career Guidance: Uses advanced language models to give personalized career advice, adapting to individual goals and career paths. Mock Interview Practice: Offers interactive mock interviews with real-time feedback to help users refine responses and boost interview confidence. Personalized Memory: Integrated LangChain’s memory capabilities, allowing the bot to recall past interactions, providing consistent, context-aware advice. Optimized Conversational Flow: Applied advanced prompt engineering for natural, engaging conversations that align with user needs.

LLM Powered Telegram Chatbot

12305695
In this project, I implemented a U-Net architecture to accurately perform brain tumor segmentation from MRI scans. The primary goal was to isolate and highlight regions of tumor tissue, aiding in diagnosis and treatment planning. U-Net, a convolutional neural network (CNN) designed specifically for biomedical image segmentation, is highly effective in tasks where precise localization is required. The model was trained and validated on a dataset of brain MRI images, achieving high accuracy in segmenting tumor regions. The use of U-Net’s encoder-decoder architecture enabled efficient feature extraction and detailed segmentation, even with a limited amount of data. The results demonstrate the potential of deep learning in medical imaging, particularly in automating complex tasks like tumor detection. For more details and to explore the code, check out the project on my Kaggle notebook.

Brain Tumour Segmentation Using UNet.

In this project, I implemented a U-Net architecture to accurately perform brain tumor segmentation from MRI scans. The primary goal was to isolate and highlight regions of tumor tissue, aiding in diagnosis and treatment planning. U-Net, a convolutional neural network (CNN) designed specifically for biomedical image segmentation, is highly effective in tasks where precise localization is required. The model was trained and validated on a dataset of brain MRI images, achieving high accuracy in segmenting tumor regions. The use of U-Net’s encoder-decoder architecture enabled efficient feature extraction and detailed segmentation, even with a limited amount of data. The results demonstrate the potential of deep learning in medical imaging, particularly in automating complex tasks like tumor detection. For more details and to explore the code, check out the project on my Kaggle notebook.

Brain Tumour Segmentation Using UNet.

13912985
This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

This project involved creating a wordpress store for jewelery brand, We created the theme the design inspiration and deployed the website for our client

Wordpress Website Development

5.0 · 7 Reviews
7 Reviews
User Avatar
Flag of

Yateesh Kumar T.

5.0

₹25,000 INR

5 days ago

Build Robust Database Pipelines

Great working with him again.

Python
Data Processing
SQL
MySQL
Database Programming

+3 more

Share

K

Former Client (inactive)

5.0

$400 USD

8 months ago

SmartAPI Trading Web App Development

It was good experience with Ahmed.

JavaScript
Python
Django
Software Architecture
PostgreSQL

+3 more

Share

User Avatar
Flag of

Abhiram K.

5.0

₹4,000 INR

10 months ago

Automate Data Extraction from Football Ticket Emails & PDFs (Gmail + Google Scripts + Drive)

Great Skills, Will hire again soon

Web Scraping
Process Automation
Google APIs

Share

A
Flag of

Ibrahim Ahmed Abdelhamid A.

5.0

€15 EUR

1 year ago

Data Science Pipeline -- 2 (archived)

Delivered the machine learning project on time with impressive precision—very professional, patient, and highly skilled. A truly collaborative freelancer with clear communication, proactive input, and seamless teamwork.

Share

N

Former Client (inactive)

5.0

$200 USD

1 year ago

Face Matching And Profile Finding

Absolutely outstanding! Ahmed is a tech wizard — delivered a complex deep-tech project on time, flawlessly, and with total professionalism. Reliable, sharp, and a joy to work with. Don’t think twice — hire him!

Web Scraping
Data Mining
Artificial Intelligence
AI Chatbot Development

+1 more

Share

Experience

Project Manager

Jan, 2023 - Jun, 2024

•

1 year, 4 months

Quantabyte

Jan, 2023 - Jun, 2024

•

1 year, 4 months

In Quantabyte I was a Project Manager.

Flag of

Islamabad, Pakistan

Jan, 2023 - Jun, 2024

•

1 year, 4 months

References

Q

quantabyte

@quantabyte

 ·

Project Manager

Working with Ahmed A. has been a great experience. He’s professional, reliable, and always brings a positive attitude to the team. His contributions made a real difference, and I truly enjoyed collaborating with him.

Education

Bahria University

2016 - 2020

•

4 years

B Sc Computer Science

Flag of

Pakistan

2016 - 2020

•

4 years

Qualifications

LangChain for LLM Application Development

2024

deeplearning.ai

Learned about the LangChain framework and its intricacies/

2024

LangChain Chat with Your Data

2024

deeplearning.ai

Learned about Retrieval Augmented Generation (RAG), a common LLM application that retrieves contextual documents from an external dataset.

2024

CS50's Introduction to Artificial Intelligence with Python

2021

Harvard University

Introductory course on how to use machine learning and artificial intelligence in Python.

2021

Publications

A SURVEY ON MACHINE LEARNING ALGORITHMS FOR QUERY BASED TEXT SUMMARIZATION

2020

ICETEMS-2020

This survey focuses on machine learning approaches for query-based text summarization, addressing the challenge of extracting relevant information efficiently from vast data sources. It reviews supervised and unsupervised learning techniques in Natural Language Processing (NLP) aimed at reducing search time and effort. Selected algorithms are analyzed based on existing literature, with detailed discussion on their internal mechanisms for effective summarization.

2020

A CASE STUDY ON GAME DEVELOPMENT USING DIFFERENT SOFTWARE MODELS WITH THREE PERSON TEAM

2018

ICETEMS : -18-082

This case study explores the development of PC and Android games by small teams using iterative and incremental SDLC models. As gaming evolves into a major global industry, efficient development with limited resources is essential. The study identifies key challenges, including tool selection, environment configuration, performance optimization, and integration testing, and provides structured recommendations to address these issues in future small-team game development projects.

2018

UTILIZATION OF FINGER BASED SENSORS IN GENERATING ECG REPORT TO PROVIDE EASE TO PATIENTS

2018

3rd ICETEMS 2018

This paper proposes a portable, finger-based ECG system that eliminates the need for traditional electrodes and gels, reducing patient discomfort and noise in readings. The system captures bio-signals from a finger in about 7 seconds, enhances them using capacitors and buffers, and transfers data via USB for filtration and noise removal. Deployed in hospitals, the prototype showed satisfactory results compared to conventional ECG machines.

2018

Certifications

Python - Level 2

Python

US English - Level 1

US English

Python Level 1

Python

Freelancer logo US (International) / English
Help & Support

Freelancer

Categories Projects Contests Freelancers Enterprise Innovation Challenges AI Development Membership Preferred Freelancer Program Project Management Local Jobs Photo Anywhere Showcase API for Developers Get Verified

About

About us How it Works Security Investor Sitemap Stories News Team Awards Press Releases Careers

Terms

Privacy Policy Terms and Conditions Copyright Policy Code of Conduct Fees and Charges

Partners

Escrow.com Loadshift Warrior Forum

Apps

Apple App Store logoGoogle Play logo

Registered Users

Total Jobs Posted

Freelancer ® is a registered Trademark of Freelancer Technology Pty Limited (ACN 142 189 759)

Copyright © 2026 Freelancer Technology Pty Limited (ACN 142 189 759)