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

Hi, I'm Monish Sri Sai Devineni—a dedicated AI/ML developer with 3+ years of hands-on experience designing, deploying, and optimizing scalable ML systems across NLP, computer vision, and predictive modeling. I focus on production-ready solutions with measurable business impact and a strong emphasis on performance, reliability, and explainability. I build end-to-end ML pipelines in Python (NumPy, Pandas, Scikit-learn) and deploy models in containerized environments using Docker, FastAPI, and Flask, leveraging transformer-based architectures (BERT, LLaMA2, GPT-4) with LangChain and vector databases for semantic search and conversational AI.…Hi, I'm Monish Sri Sai Devineni—a dedicated AI/ML developer with 3+ years of hands-on experience designing, deploying, and optimizing scalable ML systems across NLP, computer vision, and predictive modeling. I focus on production-ready solutions with measurable business impact and a strong emphasis on performance, reliability, and explainability. I build end-to-end ML pipelines in Python (NumPy, Pandas, Scikit-learn) and deploy models in containerized environments using Docker, FastAPI, and Flask, leveraging transformer-based architectures (BERT, LLaMA2, GPT-4) with LangChain and vector databases for semantic search and conversational AI.

Monish Sri Sai Devineni

AI Developer, AI Engineer, Data Scientist, +5





Hi, I'm Monish Sri Sai Devineni—a dedicated AI/ML developer with 3+ years of hands-on experience designing, deploying, and optimizing scalable ML systems across NLP, computer vision, and predictive modeling. I focus on production-ready solutions with measurable business impact and a strong emphasis on performance, reliability, and explainability. I build end-to-end ML pipelines in Python (NumPy, Pandas, Scikit-learn) and deploy models in containerized environments using Docker, FastAPI, and Flask, leveraging transformer-based architectures (BERT, LLaMA2, GPT-4) with LangChain and vector databases for semantic search and conversational AI.…Hi, I'm Monish Sri Sai Devineni—a dedicated AI/ML developer with 3+ years of hands-on experience designing, deploying, and optimizing scalable ML systems across NLP, computer vision, and predictive modeling. I focus on production-ready solutions with measurable business impact and a strong emphasis on performance, reliability, and explainability. I build end-to-end ML pipelines in Python (NumPy, Pandas, Scikit-learn) and deploy models in containerized environments using Docker, FastAPI, and Flask, leveraging transformer-based architectures (BERT, LLaMA2, GPT-4) with LangChain and vector databases for semantic search and conversational AI.

Available to hire

Hi, I’m Monish Sri Sai Devineni—a dedicated AI/ML developer with 3+ years of hands-on experience designing, deploying, and optimizing scalable ML systems across NLP, computer vision, and predictive modeling. I focus on production-ready solutions with measurable business impact and a strong emphasis on performance, reliability, and explainability.

I build end-to-end ML pipelines in Python (NumPy, Pandas, Scikit-learn) and deploy models in containerized environments using Docker, FastAPI, and Flask, leveraging transformer-based architectures (BERT, LLaMA2, GPT-4) with LangChain and vector databases for semantic search and conversational AI.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Language

English

Fluent

Work Experience

Machine Learning Engineer at Morgan Stanley

November 1, 2024 - Present

Built a high-impact credit risk prediction model using XGBoost and financial attributes to assess borrower profiles, improving loan approval accuracy by 32% and minimizing defaults. Automated feature engineering workflows using Pandas and Scikit-learn, reducing manual preprocessing by 70%. Designed anomaly detection pipeline using Isolation Forest to detect unusual patterns, decreasing fraud analyst investigation time by 45%. Developed custom financial NLP pipelines for faster report parsing, increasing speed by 39%. Created interactive Power BI dashboards for complex investment data that improved reporting efficiency by 60%. Automated compliance and regulatory report generation via Python scripting, slashing manual prep by 70%. Led A/B testing of modeling strategies, improving financial forecasting precision by 21%.

AI/ML Developer at Accenture

July 31, 2023 - July 29, 2025

Deployed multiple ML models on AWS with automated retraining, monitoring, and CI/CD pipelines, improving deployment stability by 30%. Fine-tuned BERT sentiment analysis on 100,000+ feedback entries, boosting accuracy by 22%. Integrated OpenAI GPT-3 for real-time customer log summaries, reducing response time by 41%. Developed scalable document classification pipeline using OCR and TensorFlow CNNs, cutting compliance turnaround by 50%. Built NLP pipelines for entity recognition helping marketing teams identify trends 18% faster. Designed Tableau dashboards to present sentiment and topic analysis, enabling actionable business insights at a glance.

AI/ML Engineer Intern at Accenture

July 31, 2022 - July 29, 2025

Developed customer churn prediction using Random Forest and Logistic Regression with 87% accuracy, boosting retention strategies. Built real-time anomaly detection models with OpenCV and LSTM, reducing manual video reviews by 55%. Automated hyperparameter tuning pipelines in PyTorch and TensorFlow, reducing training time by 60%. Created Power BI dashboards to visualize model accuracy and performance metrics. Engineered document parsing with CNNs and attention for 50,000+ regulatory forms monthly to cut compliance efforts by 65%. Implemented image classification with EfficientNet transfer learning, improving document recognition accuracy by 31%.

AI & Deep Learning Developer at Accenture

August 1, 2021 - July 1, 2023

Architected deep learning models for financial time-series forecasting, improving asset price prediction accuracy by 22%. Implemented NLP pipelines using BERT for sentiment analysis of financial news and social media, boosting trading signal precision by 18%. Integrated OpenAI GPT-3 with prompt engineering for real-time summaries of customer logs, reducing response times by 41%. Executed scalable, cloud-native ML workflows on Google Cloud Platform (BigQuery, Vertex AI), reducing retraining time and latency by 50%. Applied unsupervised learning and autoencoders for transaction anomaly detection, reducing false positives by 30%. Streamlined end-to-end CI/CD pipelines with Docker, Kubernetes, and GitHub Actions, enabling faster production rollouts. Collaborated with analysts, traders, and compliance to ensure regulatory KPIs and governance.

AI/ML Engineer at Accenture

June 1, 2020 - July 1, 2021

Devised deep learning models (CNN, RNN) for automated document classification and object detection, increasing operational throughput by 40%. Fine-tuned NLP models with BERT using Hugging Face for sentiment analysis and intent detection, improving classification accuracy by 20%. Built reusable PyTorch/TensorFlow pipelines with automated hyperparameter tuning and batch inference. Deployed real-time AI services using OpenCV and LSTM to detect abnormal behaviors in video streams, reducing manual fraud monitoring by 55%. Migrated legacy models to AWS SageMaker with containerized deployments and API Gateway/Lambda endpoints, cutting deployment time by 50%. Implemented model evaluation, drift detection with MLflow and CloudWatch; created Tableau dashboards showing prediction confidence and KPI impact.

AI/ML Developer at Accenture

August 1, 2022 - July 1, 2023

Deployed multiple ML models on AWS EC2 with automated retraining workflows and CI/CD, improving deployment stability by 30%. Fine-tuned a BERT-based sentiment model on 100k+ entries, increasing accuracy by 22%. Integrated OpenAI GPT-3 for real-time summaries of lengthy logs, reducing response time by 41%. Built a scalable document classification pipeline combining OCR with TensorFlow CNNs, processing 10k+ documents monthly and cutting turnaround by 50%. Developed NLP pipelines with spaCy/NLTK for entity recognition and keyword extraction, enabling faster qualitative analysis. Designed interactive Tableau dashboards to visualize NLP outputs such as sentiment drift and topic clusters.

AI/ML Engineer Intern at Accenture

June 1, 2021 - July 1, 2022

Developed churn prediction and segmentation models achieving 87% accuracy for a telecom client. Implemented real-time surveillance analytics with OpenCV and LSTM, reducing manual video reviews by 55%. Automated training pipelines in PyTorch and TensorFlow to optimize hyperparameter tuning, cutting training time by 60%. Built Power BI dashboards to monitor model performance, drift, and latency. Engineered document parsing models using CNNs and attention to extract data from 50K forms per month, reducing manual effort by 65%. Implemented EfficientNet-based image classification for smarter form processing automation.