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

Hi, I'm Himanshu Nimbarte, a passionate Machine Learning Engineer and Data Scientist currently based in Tucson, AZ. I specialize in NLP and healthcare applications, having worked on projects such as ASD detection and wheeze classification using state-of-the-art models like BioBERT and Time Delay Neural Networks. I enjoy optimizing ML pipelines and working with large datasets to uncover meaningful insights. I've had the opportunity to contribute to NIH-funded research and develop real-time anomaly detection systems, combining my skills in distributed computing, cloud technologies, and advanced ML frameworks. Whether it's fine-tuning models or automating data workflows, I love building solutions that make a tangible impact.…Hi, I'm Himanshu Nimbarte, a passionate Machine Learning Engineer and Data Scientist currently based in Tucson, AZ. I specialize in NLP and healthcare applications, having worked on projects such as ASD detection and wheeze classification using state-of-the-art models like BioBERT and Time Delay Neural Networks. I enjoy optimizing ML pipelines and working with large datasets to uncover meaningful insights. I've had the opportunity to contribute to NIH-funded research and develop real-time anomaly detection systems, combining my skills in distributed computing, cloud technologies, and advanced ML frameworks. Whether it's fine-tuning models or automating data workflows, I love building solutions that make a tangible impact.

Himanshu Nimbarte





Hi, I'm Himanshu Nimbarte, a passionate Machine Learning Engineer and Data Scientist currently based in Tucson, AZ. I specialize in NLP and healthcare applications, having worked on projects such as ASD detection and wheeze classification using state-of-the-art models like BioBERT and Time Delay Neural Networks. I enjoy optimizing ML pipelines and working with large datasets to uncover meaningful insights. I've had the opportunity to contribute to NIH-funded research and develop real-time anomaly detection systems, combining my skills in distributed computing, cloud technologies, and advanced ML frameworks. Whether it's fine-tuning models or automating data workflows, I love building solutions that make a tangible impact.…Hi, I'm Himanshu Nimbarte, a passionate Machine Learning Engineer and Data Scientist currently based in Tucson, AZ. I specialize in NLP and healthcare applications, having worked on projects such as ASD detection and wheeze classification using state-of-the-art models like BioBERT and Time Delay Neural Networks. I enjoy optimizing ML pipelines and working with large datasets to uncover meaningful insights. I've had the opportunity to contribute to NIH-funded research and develop real-time anomaly detection systems, combining my skills in distributed computing, cloud technologies, and advanced ML frameworks. Whether it's fine-tuning models or automating data workflows, I love building solutions that make a tangible impact.

Available to hire

Hi, I’m Himanshu Nimbarte, a passionate Machine Learning Engineer and Data Scientist currently based in Tucson, AZ. I specialize in NLP and healthcare applications, having worked on projects such as ASD detection and wheeze classification using state-of-the-art models like BioBERT and Time Delay Neural Networks. I enjoy optimizing ML pipelines and working with large datasets to uncover meaningful insights.

I’ve had the opportunity to contribute to NIH-funded research and develop real-time anomaly detection systems, combining my skills in distributed computing, cloud technologies, and advanced ML frameworks. Whether it’s fine-tuning models or automating data workflows, I love building solutions that make a tangible impact.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Language

English

Advanced

Work Experience

Machine Learning Engineer (Research) at University of Arizona - Deep Target NLP Lab

May 31, 2025 - July 18, 2025

Enhanced ASD detection in an NIH-funded research project by fine-tuning BioBERT on 100k clinical records, raising the F1-score by 11 percentage points. Reduced model attention drift by 12% through customized attention heads and domain-specific preprocessing. Applied clustering algorithms to uncover latent topics in clinical narratives, evaluating cluster quality with weighted Z-scores. Achieved a 90% reduction in model training time by parallelizing 10-fold cross-validation using SLURM on HPC GPU clusters.

Student Researcher (Machine Learning) at University of Arizona

August 31, 2024 - July 18, 2025

Improved wheeze classification accuracy by 5.2% via decision threshold optimization and self-supervised learning feature extraction using an Enhanced TDNN model in PyTorch. Enhanced sound analysis pipeline through bandpass filtering and Mel-spectrogram visualization. Developed structured datasets from audio using Wav2Vec2 embeddings for efficient ML training.