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

I'm Rajat Balda, an AI/ML Engineer with 4+ years of experience building end-to-end ML solutions across healthcare and analytics. I specialize in multi-modal modeling with medical imaging, tabular data, and NLP, and I work with CNNs, Transformers, gradient-boosting models, and ML tooling like MLflow and TorchScript/ONNX. I have a proven track record in data pipeline engineering, model optimization, and deploying ML systems via REST APIs to deliver reliable, production-ready insights. I enjoy turning complex data into actionable business outcomes through clear visualization, robust evaluation, and explainability using Grad-CAM and SHAP.…I'm Rajat Balda, an AI/ML Engineer with 4+ years of experience building end-to-end ML solutions across healthcare and analytics. I specialize in multi-modal modeling with medical imaging, tabular data, and NLP, and I work with CNNs, Transformers, gradient-boosting models, and ML tooling like MLflow and TorchScript/ONNX. I have a proven track record in data pipeline engineering, model optimization, and deploying ML systems via REST APIs to deliver reliable, production-ready insights. I enjoy turning complex data into actionable business outcomes through clear visualization, robust evaluation, and explainability using Grad-CAM and SHAP.

Rajat Balda

AI Engineer, Data Scientist, Full Stack Developer, +4





I'm Rajat Balda, an AI/ML Engineer with 4+ years of experience building end-to-end ML solutions across healthcare and analytics. I specialize in multi-modal modeling with medical imaging, tabular data, and NLP, and I work with CNNs, Transformers, gradient-boosting models, and ML tooling like MLflow and TorchScript/ONNX. I have a proven track record in data pipeline engineering, model optimization, and deploying ML systems via REST APIs to deliver reliable, production-ready insights. I enjoy turning complex data into actionable business outcomes through clear visualization, robust evaluation, and explainability using Grad-CAM and SHAP.…I'm Rajat Balda, an AI/ML Engineer with 4+ years of experience building end-to-end ML solutions across healthcare and analytics. I specialize in multi-modal modeling with medical imaging, tabular data, and NLP, and I work with CNNs, Transformers, gradient-boosting models, and ML tooling like MLflow and TorchScript/ONNX. I have a proven track record in data pipeline engineering, model optimization, and deploying ML systems via REST APIs to deliver reliable, production-ready insights. I enjoy turning complex data into actionable business outcomes through clear visualization, robust evaluation, and explainability using Grad-CAM and SHAP.

Available to hire

I’m Rajat Balda, an AI/ML Engineer with 4+ years of experience building end-to-end ML solutions across healthcare and analytics. I specialize in multi-modal modeling with medical imaging, tabular data, and NLP, and I work with CNNs, Transformers, gradient-boosting models, and ML tooling like MLflow and TorchScript/ONNX.

I have a proven track record in data pipeline engineering, model optimization, and deploying ML systems via REST APIs to deliver reliable, production-ready insights. I enjoy turning complex data into actionable business outcomes through clear visualization, robust evaluation, and explainability using Grad-CAM and SHAP.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Language

English

Fluent

Work Experience

AI/ML Engineer at Abbott Laboratories

November 1, 2024 - Present

Designed and implemented a full multi-modal ML pipeline integrating DICOM medical imaging with lab-test and patient-history data for combined predictive diagnostics. Built and optimized CNN-based models for anomaly detection and fused them with gradient-boosting tabular models using a custom late-fusion architecture. Engineered a robust data-processing system for heterogeneous medical datasets, implementing automated handling of missing values, OOD samples, and corrupted images. Implemented model-explainability components using Grad-CAM and SHAP to provide clinically interpretable insights. Conducted rigorous model evaluation using AUC-ROC, precision–recall, sensitivity/specificity, calibration curves, and cross-validation. Developed and deployed an end-to-end diagnostic prototype with REST APIs, packaging models using TorchScript/ONNX, and built a UI for risk-score visualization. Established experiment-tracking and versioning workflows using MLflow/W&B, documented key decisions, and

ML Engineer at Infinite Infolab

August 1, 2020 - July 1, 2023

Developed scalable classification and forecasting models using Scikit-learn, XGBoost, and Python, improving client decision accuracy by 35% and accelerating insight generation for key workflows. Built CNN-based defect-detection pipelines with OpenCV, increasing detection precision from 70% to 92% and reducing quality-control losses for manufacturing clients. Engineered NLP systems using Hugging Face Transformers and BERT for summarization and intent analysis, eliminating manual review workloads and enabling automated text-processing at scale. Designed and optimized high-volume data pipelines using Pandas, NumPy, and SQL, cutting data processing time by 50% across multimillion-row datasets. Delivered analytics dashboards in Tableau and Power BI that translated model outputs into actionable business insights, improving stakeholder decision speed and clarity. Standardized reproducible ML environments using Docker and Git, reducing onboarding overhead and deployment friction across enginee