Adarsh Singh

AI Researcher

Greater Delhi, India4 yrs 7 mos experience

AI EnabledAI ML Practitioner

Key Highlights

5+ years of experience in data science and machine learning.
Expertise in fraud detection and financial modeling.
Proven track record in developing AI-based solutions.

Stackforce AI infers this person is a Data Scientist specializing in Fintech and AI-driven solutions.

Contact

adarshksr@gmail.com LinkedIn

Skills

Core Skills

Statistical ModelingMachine LearningFraud DetectionNatural Language ProcessingData EngineeringData ScienceComputer VisionArtificial Intelligence

Other Skills

AI-based SolutionsBusinessClassification AlgorithmsCloudCredit RiskData CollectionData ValidationData structures & AlgorithmsDeep LearningETL Process AutomationEconomicsFace Liveness DetectionFinancial MarketsFraud Models DevelopmentGenerative AI

About

Bachelor of Technology graduate from the Indian Institute of Technology (IIT), Roorkee. Currently working as a Quants Data Scientist at Barclays. 5+ years of experience solving business problems using Python, Machine Learning, and Statistics. I have worked on Quantitative Modelling, Financial Modeling, Fraud Detection, Anomaly Detection, Advertisement Classification, Toxic Classification, Named Entity Recognition, Unsupervised Machine Learning and Insurance Clause Similarities, End-to-End Model Development and Deployment, Model Monitoring and Model Validation. I have worked on fine-tuning of Bert Model and other Large Language Models. Considerable work experience with Keras and Tensorflow Framework. Skilled in Python, Machine Learning, Deep Learning, Neural Networks, Natural Language Processing, Time Series Analysis, Fraud Detection, Credit Risk, Statistical Modelling, Predictive Modelling, Generative AI, LLM, BERT, GPT, Unsupervised ML Techniques, Clustering, Computer Vision, SQL, SAS, PyTorch, PySpark, Cloud, Data structures & Algorithms, Probability, Statistics, Financial Markets, Business, and Economics.

Experience

Barclays

Data Scientist (Quants)

Apr 2025 – Present · 11 mos

Quant Model Developer
Market Risk
Credit Risk
Regression
Statistical Modelling
Investment Banking
Stress Testing
RWA
Balance Sheets
Assets & Liabilities Forecasting
PPNR
Regulatory Risk
CCAR, PRA

Quant Model DevelopmentMarket RiskCredit RiskRegressionStatistical ModellingInvestment Banking+3

Bank of america

Data Scientist

Nov 2022 – Mar 2025 · 2 yrs 4 mos

Fraud Detection
Fraud Models Development
Fraud Models Validation and Monitoring
Statistical Modeling
Unsupervised Machine Learning
Natural Language Processing
Generative AI
LLM

Fraud DetectionFraud Models DevelopmentStatistical ModelingUnsupervised Machine LearningNatural Language ProcessingGenerative AI+2

Ltimindtree

Data Science Engineer

Jul 2021 – Nov 2022 · 1 yr 4 mos · Mumbai, Maharashtra, India · Remote

Modeled Sensitive Data Discovery (Named Entity Recognition) for Financial and Healthcare Organizations to recognize sensitive information such as Names, Addresses, E-mail Addresses, Contact numbers, Bank Account numbers, Credit Card numbers etc., accounting for 95% cost savings compared to manual efforts of PII Identification
Automated and Integrated the ETL process to save and transform the outcomes in specific file formats
Created an AI-based solution for the policymakers to analyze the similarities between clauses of Insurance Domain and group similar ones within the same categories that resulted in a financial impact of ~$5MM
Involved in Data Validation, Query Optimization along with product owners and bug fixes for raised issues
Refined automated Stored procedures for Govt. of India ‐ saved tens of hours per employee in having to manually load the data and perform the Growth Rate, CAGR and several other calculations and numerous validations every year

Named Entity RecognitionETL Process AutomationAI-based SolutionsData ValidationQuery OptimizationNatural Language Processing+1

Jungleworks

Artificial Intelligence Engineer

May 2020 – Jul 2020 · 2 mos

Research and Development on Face Liveness Detection techniques using images and videos from a 2-D camera
Developed a Face Liveness Detection module using Local Binary Patterns, which detects the liveness with an accuracy of 80% for ~2k employees and more than 10k Cab and Auto Drivers which accounts for a financial impact of ~$2MM
After attaining a suitable optimization between speed and accuracy, deployed it and assisting more than 10k users

Face Liveness DetectionImage ProcessingModule DevelopmentComputer VisionArtificial Intelligence

Spacenos

Natural Language Processing Engineer

Apr 2020 – Jun 2020 · 2 mos

Implemented LSTM, which was later converted into tflite for an app that detects the percentage of toxicity, hate, insult, and four more categories in a given sentence. The accuracy achieved was 93%
Modeled a Question Pairs Dataset from Quora which identifies the extent of similarity between two sentences
Brainstormed on pre-existing several Fake news and Rumour Detection techniques in textual format
Engaged with the developer team and the Market Research team to be well acquainted with whatever they require

Toxicity DetectionLSTM ImplementationQuestion Pairs Dataset ModelingNatural Language ProcessingMachine Learning

Samsung electronics

Machine Learning Researcher

Dec 2019 – Aug 2020 · 8 mos · IIT Roorkee , India · On-site

The task was to upgrade Samsung Antivirus and make it able to identify new Malware apps.
Collected API calls data using various Samsung smartphones with pre-installed malicious and benign apps.
Implemented different classification algorithms like RandomForest, XGBoost, and ANN for developing the model after using dimension reduction techniques (PCA) to accumulate the maximum variance for detecting the malware
Used the F-1 score and Confusion Matrix for evaluating the performance of the model, and the maximum F-1 score achieved was 0.95 for an unbalanced dataset and 0.90 for a balanced dataset, conceived using the ANN algorithm.
Identified 46 out of 50 malware correctly using the ultimate classifier versus 30 out of 50 by the Samsung antivirus

Malware DetectionClassification AlgorithmsData CollectionModel EvaluationMachine LearningData Science