Sayali Kandarkar
Mountain View, California, United States
2K followers
500 connections
About
My interests lie at the intersection of technology and entrepreneurship.
I strongly…
Activity
-
Woke up to a frenzied internet today due to the CrowdStrike outage. Having been working on large-scale software rollouts over the past few months I…
Woke up to a frenzied internet today due to the CrowdStrike outage. Having been working on large-scale software rollouts over the past few months I…
Liked by Sayali Kandarkar
-
We’re excited to announce that our CEO Jensen Huang will be joined by Meta CEO Mark Zuckerberg to discuss how fundamental research is enabling AI…
We’re excited to announce that our CEO Jensen Huang will be joined by Meta CEO Mark Zuckerberg to discuss how fundamental research is enabling AI…
Liked by Sayali Kandarkar
-
We’re striving to build the most scalable, robust, and generalizable robotics foundation model. Thank you to our team for all their hard work so far…
We’re striving to build the most scalable, robust, and generalizable robotics foundation model. Thank you to our team for all their hard work so far…
Liked by Sayali Kandarkar
Experience
Education
-
Carnegie Mellon University School of Computer Science
Activities and Societies: Graduate Student Assembly Representative (LTI department) Women in Computer Science (WiCS) Women in Machine Learning (WiML)
Courses -
Spring 2024
11-699: MSAII Capstone (Working on building luca.ai for dyslexic students)
10-605: Machine Learning for Large Datasets
54-192: Acting Ensemble for Non-Majors
Fall 2023
11-667: Large Language Models, Methods and Applications
11-711: Advanced Natural Language Processing
11-777: Multimodal Machine Learning
11-654: AI Innovation
69-150: Swimming
Spring 2023
11-785: Introduction to Deep Learning
11-695: AI Engineering /…Courses -
Spring 2024
11-699: MSAII Capstone (Working on building luca.ai for dyslexic students)
10-605: Machine Learning for Large Datasets
54-192: Acting Ensemble for Non-Majors
Fall 2023
11-667: Large Language Models, Methods and Applications
11-711: Advanced Natural Language Processing
11-777: Multimodal Machine Learning
11-654: AI Innovation
69-150: Swimming
Spring 2023
11-785: Introduction to Deep Learning
11-695: AI Engineering / Machine Learning in Production
45-996: Corporate Startup Lab
11-636: Independent Study (Mental Health in AI at Stanford)
Fall 2022
11-601: Coding Boot Camp
10-601: Introduction to Machine Learning
11-651: Artificial Intelligence in Future Markets
17-762: Law in Computer Technology
Summer 2022
15-513: Introduction to Computer Systems -
I did my under graduation in Information Technology Engineering. I learned various computer fundamentals here like Operating Systems, DBMS, Computer Networks, Big Data Analytics and much more. I took those fundamentals and implemented the knowledge gained by academics into various demonstrative projects.
-
-
State Board Topper
Licenses & Certifications
-
Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization
Coursera
Issued
Volunteer Experience
-
Software Developer/Volunteer
India Shield
- 6 months
Health
- Developed a python bot to scrape tweets comprising available covid resources like oxygen, ICU beds, plasma etc; performed pre-processing using NLTK, spacy
- Integrated the data with AWS; enabled dynamic API calls with the help of Amazon API Gateway, AWS Lambda, and DynamoDB to store data and update covid website (https://warwithcovid.club/) on runtime, thus helping 100k Indian citizens meet urgent needs; featured in Forbes India Issue for the same. -
Subject Matter Expert
JPMorgan Chase & Co.
- 1 year
Civil Rights and Social Action
- Exhibited technical expertise to guide external candidates to overcome challenges they faced as part of the Code for Good hackathon; I served as a “brand ambassador” for the firm inspiring the next class of software engineers.
- Led multiple teams of 6 candidates each through project ideation to execution during Code for Good hackathon; thus, helping them build innovative AI solutions for 40 non-profit partners.
Publications
-
Preterm Birth Detection using Convolutional Neural Networks
According to the World Health Organization, approximately 1 million children die each year due to preterm birth complications (1). Many survivors face a lifetime of disability, including learning disabilities and visual and hearing problems. Inequalities in survival rates around the world are stark. In low-income settings, half of the babies born at or below 32 weeks (2 months early) die due to a lack of feasible, cost-effective care, such as warmth, breastfeeding support, and primary care for…
According to the World Health Organization, approximately 1 million children die each year due to preterm birth complications (1). Many survivors face a lifetime of disability, including learning disabilities and visual and hearing problems. Inequalities in survival rates around the world are stark. In low-income settings, half of the babies born at or below 32 weeks (2 months early) die due to a lack of feasible, cost-effective care, such as warmth, breastfeeding support, and primary care for infections and breathing difficulties. India is the country with the most significant number of preterm births (2). One of the reasons could be the lack of doctors to examine the medical conditions of pregnant women manually. Over 44% of WHO Member States in India reported less than one physician per 1,000 population (3). Hence, developing countries such as India need Artificial Intelligence (AI) systems that would assist the clinicians and eventually require less to no manual intervention to predict such conditions. Several studies have reported that cervical assessment on transvaginal sonography may help predict preterm delivery (4,5). In this paper, a segmentation technique is proposed to predict preterm birth using Convolutional Neural Networks (CNNs). The proposed model is trained on a dataset of 1334 transvaginal ultrasonic 2D images. According to the result, the U-net-based CNN approach achieved more promising results than the current state-of-the-art techniques.
-
Brain Tumor Segmentation Using U-NET Based Convolutional Neural Network
International Journal of Innovative Research in Computer and Communication Engineering
Brain tumor segmentation can be used to separate healthy tissues from cancerous tissues accurately.
Usually, in clinical settings, a brain MRI (Magnetic Resonance Imaging) scan is used to detect
tumors, but manually segmenting out a tumor is a time-consuming process and requires highly skilled
and experienced radiologists. If a malignant tumor is not detected in the early stages of cancer, it can
cost a life. Hence, early detection of tumors in a short amount of time would be a…Brain tumor segmentation can be used to separate healthy tissues from cancerous tissues accurately.
Usually, in clinical settings, a brain MRI (Magnetic Resonance Imaging) scan is used to detect
tumors, but manually segmenting out a tumor is a time-consuming process and requires highly skilled
and experienced radiologists. If a malignant tumor is not detected in the early stages of cancer, it can
cost a life. Hence, early detection of tumors in a short amount of time would be a great boon to
humanity. This paper proposes a brain tumor segmentation technique using a Deep Convolutional
Neural Network (CNN). The type of Convolutional Neural Network proposed is called U-NET. The
U-NET architecture uses semantic segmentation that labels every pixel of an image rather than just
detecting objects. So, if we can segment out an image, we would know which pixel belongs to what
part of a human’s anatomy. It can help us efficiently detect all types of tumors and irregularities,
which would help the radiologists and surgeons with detection and surgery. -
A Survey on Clickjacking and Tapjacking Solutions Provided by Different Browsers
International Journal of Innovative Research in Computer and Communication Engineering
Projects
-
AWS DeepRacer
- Present
Developed a reinforcement model to train, evaluate and tune a 3-layer convolutional neural network controlling a virtual 1/18th scale race car.
-
Brain tumor segmentation using U-NET based deep CNN
-
Trained a model based on U-net architecture that comprised of 11 layers(conv2d, ReLU, max-pooling) for both encoding as well as decoding.
U-NET incorporates segmentation technique where every pixel of an image is labeled in order to be accurately classified. The architecture usually has 2 parts - encoding(normal convolutions) and decoding(transpose convolutions). Transpose convolutions are used to gain the spatial information of the image by pulling activations from previous building…Trained a model based on U-net architecture that comprised of 11 layers(conv2d, ReLU, max-pooling) for both encoding as well as decoding.
U-NET incorporates segmentation technique where every pixel of an image is labeled in order to be accurately classified. The architecture usually has 2 parts - encoding(normal convolutions) and decoding(transpose convolutions). Transpose convolutions are used to gain the spatial information of the image by pulling activations from previous building blocking using 'skip-connections'.
The model was trained on the 'fignet' website dataset of 3024 MRI scans. The model obtained an accuracy of 95.7% -
Transfer Learning based CNN classifier
-
Sometimes, we don't have enough computational power or the resources to build highly deep neural networks. In that case, we can use the already built neural networks; freeze the weights; replace the sigmoid function as per our need and then run the model to classify!
In this project, I’ve used the existing MobileNetV2 architectural CNN as a pre-trained network and built a classifier to detect if the image contains an alpaca or not -
COVID-19 Twitter Bot
-
Developed a python bot to scrape tweets comprising available covid resources like oxygen, ICU beds, plasma etc; performed pre-processing using NLTK, spacy
Integrated the data with AWS; enabled dynamic API calls with the help of Amazon API Gateway, AWS Lambda, and DynamoDB to store data and update covid website (https://warwithcovid.club/) on runtime, thus helping 100k Indian citizens meet urgent needs; featured in Forbes India Issue for the same. -
RASA chatbot
-
Developed a chatbot using NLTK and NLP for text analysis, summarisation and lemmatisation! Further utilised RASA framework and added intents and entities for text classification; integrated it with the Springboot REST APIs to send the response back to the clients!
-
Customer Churn Analysis
-
Built a ML model using Random Forest Classified and XGBoost to predict the customer churn rate; thus helping the firm take measures for retention of customers
-
Breast Cancer Detection System
-
Consulted with pathologists to understand breast cancer diagnosis; formulated 3 critical parameters to be extracted using image segmentation pipelines.
Devised a novel model pipeline to segment tumor cells; employed novel U-Net & OpenCV tools to generate semantic & instance segmentation; obtained an improvement of 6% in DICE score.
Honors & Awards
-
Certificate of Recognition
J P Morgan Chase & Co
Certificate of recognition for leading/volunteering at multiple GoodWorks events.
-
Certificate of Merit
K J Somaiya College of Engineering
Secured highest marks in a course in the 4th year of Information technology engineering.
-
Government of Maharashtra Merit Award
Mumbai police
Secured highest percentage(100%) throughout Maharashtra in the Secondary School Certificate examination.
-
17th Senior State Softball Championship
Government of Maharashtra
Certificate of participation by Maharashtra state softball association for participation in the 17th senior state softball championship held at Garware stadium, Aurangabad
Languages
-
English
Native or bilingual proficiency
-
Hindi
Native or bilingual proficiency
-
Marathi
Native or bilingual proficiency
-
German
Limited working proficiency
More activity by Sayali
-
We're excited to share that Gemma 2, the next gen of Google Gemma models, is now optimized with TensorRT-LLM and packaged as NVIDIA NIM inference…
We're excited to share that Gemma 2, the next gen of Google Gemma models, is now optimized with TensorRT-LLM and packaged as NVIDIA NIM inference…
Liked by Sayali Kandarkar
-
NVIDIA just became the World’s most valuable company. https://lnkd.in/eTx4TNeB
NVIDIA just became the World’s most valuable company. https://lnkd.in/eTx4TNeB
Liked by Sayali Kandarkar
-
⚡ NVIDIA Wins #CVPR2024 Autonomous Grand Challenge for End-to-End Driving NVIDIA Research topped the CVPR leaderboard in the end-to-end driving at…
⚡ NVIDIA Wins #CVPR2024 Autonomous Grand Challenge for End-to-End Driving NVIDIA Research topped the CVPR leaderboard in the end-to-end driving at…
Liked by Sayali Kandarkar
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More