Sagarika Jadon

AI Researcher

Bengaluru, Karnataka, India3 yrs 1 mo experience

AI EnabledAI ML Practitioner

Key Highlights

Skilled in machine learning and deep learning with practical experience.
Proven track record in developing AI models for real-world applications.
Strong programming skills in Python, C, and C++.

Stackforce AI infers this person is a Machine Learning Specialist with expertise in AI-driven solutions.

Contact

Skills

Other Skills

AI ExplainabilityAirflowAlgorithmsAnalytical SkillsBERTCC++CaptumCatboostComputer VisionConvolutional Neural Networks (CNN)Data AnalysisData AnalyticsData ScienceData Structures

About

With close to 2 years of experience, I am a skilled Machine Learning Scientist, specialising in machine learning, deep learning, computer vision, natural language processing, and MLOps. I possess strong programming skills in Python, C, and C++, along with expertise in PyTorch, TensorFlow, Hugging Face, AWS, SQL, and more. Committed to continuous learning, I stay updated with the latest advancements in AI and ML, ensuring that I bring cutting-edge knowledge and skills to every project.

Experience

Wayfair

Machine Learning Scientist

Aug 2024 – Present · 1 yr 7 mos · Bengaluru, Karnataka, India · On-site

Part of the Customer Understanding Science team
Building GenAI pipelines for recommendations and personalisation

Hyperverge

2 roles

Machine Learning Engineer

Jul 2023 – Aug 2024 · 1 yr 1 mo · Bengaluru, Karnataka, India · On-site

Finetuned a multi-modal contrastive model to detect display attacks involving ID documents from
different geographical regions. Achieved less than 1% FAR and FRR

Deep Learning Intern

Aug 2022 – Jan 2023 · 5 mos · Bengaluru, Karnataka, India

Deepfakes Detection-
Trained a SRM convolution based deep learning model to detect image deepfakes and forged facial images for an Indonesian banking institution.
Improved the FAR and FRR of the existing model by 30% and 20% respectively.
Performed extensive literature survey on deepfakes and forged images detection.
Occlusion Detection-
Worked on the task of detecting occluded faces during customer onboarding and KYC.
Collected relevant data and managed the annotation process. Created a dataset of ~120K images.
Finetuned Resnet50 model, which achieved accuracy >99%
Passport Classification-
Finetuned MobilenetV2 on backgrounds segmented using DeeplabV3 to identify passport images.
The model achieved FAR and FRR of less than 0.5%

Shl

AI Research Intern

May 2022 – Jul 2022 · 2 mos · Gurugram, Haryana, India

Worked on SHL's SIOD (Smart Interview On Demand) product
Finetuned Bert based pretrained language models to score candidates’ responses to situational and behavioral based interview questions.
Used integrated gradients and Shapely values for interpreting the models and their outputs.