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

Hi, I’m Abhishek Reddy, an AI/ML Engineer who thrives on turning large-scale data into reliable, production-ready models and retrieval-based systems. I enjoy building high-impact solutions across customer risk modeling, document intelligence, and real-time workflows, and I’m passionate about improving model accuracy, system reliability, and decision speed in enterprise settings. I specialize in model development, RAG-based architectures, and experimentation frameworks, and I design end-to-end ML pipelines with strong emphasis on validation, monitoring, and scalable deployment to keep business decisions fast and well-informed.…Hi, I’m Abhishek Reddy, an AI/ML Engineer who thrives on turning large-scale data into reliable, production-ready models and retrieval-based systems. I enjoy building high-impact solutions across customer risk modeling, document intelligence, and real-time workflows, and I’m passionate about improving model accuracy, system reliability, and decision speed in enterprise settings. I specialize in model development, RAG-based architectures, and experimentation frameworks, and I design end-to-end ML pipelines with strong emphasis on validation, monitoring, and scalable deployment to keep business decisions fast and well-informed.

Abhishek Reddy

AI Engineer, Data Scientist, AI Developer, +4





Hi, I’m Abhishek Reddy, an AI/ML Engineer who thrives on turning large-scale data into reliable, production-ready models and retrieval-based systems. I enjoy building high-impact solutions across customer risk modeling, document intelligence, and real-time workflows, and I’m passionate about improving model accuracy, system reliability, and decision speed in enterprise settings. I specialize in model development, RAG-based architectures, and experimentation frameworks, and I design end-to-end ML pipelines with strong emphasis on validation, monitoring, and scalable deployment to keep business decisions fast and well-informed.…Hi, I’m Abhishek Reddy, an AI/ML Engineer who thrives on turning large-scale data into reliable, production-ready models and retrieval-based systems. I enjoy building high-impact solutions across customer risk modeling, document intelligence, and real-time workflows, and I’m passionate about improving model accuracy, system reliability, and decision speed in enterprise settings. I specialize in model development, RAG-based architectures, and experimentation frameworks, and I design end-to-end ML pipelines with strong emphasis on validation, monitoring, and scalable deployment to keep business decisions fast and well-informed.

Available to hire

Hi, I’m Abhishek Reddy, an AI/ML Engineer who thrives on turning large-scale data into reliable, production-ready models and retrieval-based systems. I enjoy building high-impact solutions across customer risk modeling, document intelligence, and real-time workflows, and I’m passionate about improving model accuracy, system reliability, and decision speed in enterprise settings.

I specialize in model development, RAG-based architectures, and experimentation frameworks, and I design end-to-end ML pipelines with strong emphasis on validation, monitoring, and scalable deployment to keep business decisions fast and well-informed.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Language

English

Fluent

Work Experience

AI/ML Engineer at JPMorgan Chase & Co.

January 1, 2025 - Present

Developed and deployed ML models for customer churn prediction and lead scoring on 10M+ customer records; integrated outputs into CRM workflows, improving outreach prioritization precision by 18%. Implemented RAG pipelines using vector search and contextual reranking to retrieve relevant financial documents for LLM responses; reduced factual inconsistencies in compliance-related queries by 32% and decreased analyst lookup time to under 2 minutes. Designed prompt chaining workflows for GPT-based applications, introduced validation checks and conducted A/B tests; improved task completion rates by 22% and reduced follow-up clarification requests by 35%. Built experimentation pipelines using statistical testing (t-tests, bootstrap confidence intervals) to evaluate model and prompt performance, enabling faster decision-making and reducing validation cycle time from 3 weeks to 8 days across use cases. Deployed and managed ML pipelines on GCP Vertex AI, supporting batch inference and model ve

Data Scientist at Coforge, India

August 1, 2021 - July 1, 2023

Engineered data pipelines and feature engineering workflows to support ML models for demand forecasting, processing 2M+ supply chain records per batch across 6 business units; improved data quality and reduced preprocessing time by 60%, enabling more reliable model training. Designed and maintained a centralized feature repository in SQL Server, standardizing feature definitions across 4+ predictive models and reducing redundant data transformations by 45%. Developed demand forecasting models using LSTM and XGBoost ensembles across 15+ distribution nodes, improving 30-day demand prediction accuracy by 17% (MAPE) and reducing overstock incidents by 12%. Delivered Power BI dashboards to monitor model performance and forecast accuracy, enabling 20+ stakeholders to track ML-driven decisions and reducing ad-hoc analytical requests by 50%. Collaborated with operations and procurement teams to align forecasting models with inventory planning strategies.

Data Scientist at Coforge

August 1, 2021 - July 31, 2023

Engineered data pipelines and feature engineering workflows to support ML models for demand forecasting, processing 2M+ supply chain records per batch across 6 business units; improved data quality and reduced preprocessing time by 60%, enabling more reliable model training. Designed and maintained a centralized feature repository in SQL Server, standardized feature definitions across 4+ predictive models, reducing redundant data transformations by 45%. Developed demand forecasting models using LSTM and XGBoost ensembles across 15+ distribution nodes; improved 30-day demand prediction accuracy by 17% (MAPE) and reduced overstock incidents by 12%. Evaluated forecasting models using cross-validation, backtesting, and error analysis, improving generalization. Delivered Power BI dashboards to monitor model performance, forecast accuracy, and anomaly patterns, enabling 20+ stakeholders to track ML-driven decisions and reducing ad-hoc analytical requests by 50%. Partnered with operations and