Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI & ML Engineer with 5+ years of experience designing, building, and deploying production-grade Machine Learning, Deep Learning, and Generative AI solutions across FinTech and enterprise platforms. Expertise in Predictive Analytics, Fraud Detection, Time-Series Forecasting, LLMs, RAG, NLP, MLOps, and Feature Engineering, delivering scalable AI systems on AWS and Azure. Proven track record of measurable business impact, including 44% reduction in chargebacks, 96% AUC-ROC fraud detection accuracy, and <8% MAPE cashflow forecasting, through end-to-end model development, deployment, monitoring, and continuous optimization.…AI & ML Engineer with 5+ years of experience designing, building, and deploying production-grade Machine Learning, Deep Learning, and Generative AI solutions across FinTech and enterprise platforms. Expertise in Predictive Analytics, Fraud Detection, Time-Series Forecasting, LLMs, RAG, NLP, MLOps, and Feature Engineering, delivering scalable AI systems on AWS and Azure. Proven track record of measurable business impact, including 44% reduction in chargebacks, 96% AUC-ROC fraud detection accuracy, and <8% MAPE cashflow forecasting, through end-to-end model development, deployment, monitoring, and continuous optimization.

Dhruvisha Navdipkumar Jaiswal

AI Engineer, Data Scientist, Software Engineer, +2





AI & ML Engineer with 5+ years of experience designing, building, and deploying production-grade Machine Learning, Deep Learning, and Generative AI solutions across FinTech and enterprise platforms. Expertise in Predictive Analytics, Fraud Detection, Time-Series Forecasting, LLMs, RAG, NLP, MLOps, and Feature Engineering, delivering scalable AI systems on AWS and Azure. Proven track record of measurable business impact, including 44% reduction in chargebacks, 96% AUC-ROC fraud detection accuracy, and <8% MAPE cashflow forecasting, through end-to-end model development, deployment, monitoring, and continuous optimization.…AI & ML Engineer with 5+ years of experience designing, building, and deploying production-grade Machine Learning, Deep Learning, and Generative AI solutions across FinTech and enterprise platforms. Expertise in Predictive Analytics, Fraud Detection, Time-Series Forecasting, LLMs, RAG, NLP, MLOps, and Feature Engineering, delivering scalable AI systems on AWS and Azure. Proven track record of measurable business impact, including 44% reduction in chargebacks, 96% AUC-ROC fraud detection accuracy, and <8% MAPE cashflow forecasting, through end-to-end model development, deployment, monitoring, and continuous optimization.

Available to hire

AI & ML Engineer with 5+ years of experience designing, building, and deploying production-grade Machine Learning, Deep Learning, and Generative AI solutions across FinTech and enterprise platforms. Expertise in Predictive Analytics, Fraud Detection, Time-Series Forecasting, LLMs, RAG, NLP, MLOps, and Feature Engineering, delivering scalable AI systems on AWS and Azure.

Proven track record of measurable business impact, including 44% reduction in chargebacks, 96% AUC-ROC fraud detection accuracy, and <8% MAPE cashflow forecasting, through end-to-end model development, deployment, monitoring, and continuous optimization.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Beginner

Language

Work Experience

AI & ML Engineer at Xero

March 1, 2026 - Present

Designed, trained, and deployed time-series forecasting models using LSTM to generate 30/60/90/180-day cashflow forecasts leveraging Snowflake, dbt, and engineered features (rolling statistics, seasonality, payment behavior, payroll cycles). Achieved <8% MAPE (30-day) and <15% MAPE (180-day). Built hierarchical Bayesian forecasting for cold-start/sparse-data organizations to expand coverage across 4.6M+ SMB organizations. Implemented anomaly detection pipelines using Isolation Forest and Z-score analysis to identify unusual trends (cashflow, gross profit, burn rate, accounts receivable). Developed an LLM-based insights platform using RAG (LangChain, FAISS) for contextual cashflow explanations. Implemented MLOps monitoring with MLflow for evaluation, backtesting, A/B testing, and drift/quality tracking (MAPE, precision/recall, forecast coverage).

AI & ML Engineer at Razorpay

July 1, 2022 - December 31, 2025

Developed an AI-powered Smart Payment Routing Engine using ML models (Random Forest, Logistic Regression) and low-latency inference with AWS EKS and ONNX Runtime, improving payment success rates by 4–7% and supporting 5,000+ TPS. Built real-time fraud detection models including XGBoost and Graph Neural Networks (GraphSAGE) with PyTorch Geometric, using Spark-based feature engineering, reducing chargebacks by 44% and improving fraud detection accuracy. Engineered scalable ML infrastructure on AWS using Kafka (MSK), Redis, EMR, and S3 for billions of transaction records. Supported enterprise MLOps using MLflow, Airflow, Great Expectations, drift detection, CI/CD, and automated retraining to improve governance and reliability. Collaborated across engineering, risk, and product teams to strengthen payment security and improve inference performance using cloud-native ML pipelines and streaming data processing.

AI & ML Engineer at Accenture

December 1, 2020 - July 31, 2022

Developed production ML solutions for fraud detection and customer churn prediction using Python, PySpark, and Azure Databricks, processing 6M+ daily transactions and 80M+ customer records. Built and optimized models using XGBoost, LightGBM, TensorFlow, and Scikit-learn with hyperparameter tuning, achieving 96% AUC-ROC for fraud and 89% AUC-ROC for churn. Deployed real-time inference pipelines (FastAPI, Docker, Kubernetes/AKS) at 800+ TPS and automated model training/validation/deployment using Airflow and MLflow. Implemented MLOps monitoring, drift detection, and explainability with SHAP using Grafana and PagerDuty. Delivered business outcomes including reduced fraud false positives by 76%, improved retention campaign effectiveness by 23%, and data-driven decision support.