Rishabh Singh
# [email protected] | (857)701-5831 | ï linkedin.com/in/RobuRishabh
www.RobuRishabh.com | § github.com/RobuRishabh | ü medium.com/@RobuRishabh
Experience
Computer Vision Research Assistant - Northeastern University, Boston, USA Jul 2023 - Dec 2024
{ Preprocessed 256 GB RAW image data, staged debayering, color balancing to generate 16-bit TIFF Linear/Log RGB.
{ Trained Faster R-CNN object detection model with ResNet50 and MobileNetV2 backbone, achieving mAP 0.91 on 6 classes.
{ Formulated custom augmentations and optimized PyTorch training for 20,000-image dataset, enabling real-time object detection.
Algorithms & Machine Learning Engineer - Dynocardia Inc, Boston, USA Jan 2024 - Aug 2024
{ Conducted data analysis, preprocessing, segmentation, and classification using XGBoost, CNN, Vision Transformer (ViT) on
optical flow/variance images, achieving correlation coefficients of 0.85 and 0.81 for 2 classes (Splitting and budding).
{ Attained 84.9% accuracy in classifying, tracked biomedical shapes evolution with 3 class labels, integrated SHAP feature analysis.
{ Developed a dual IMU sensors-based algorithm to estimate heart-to-wrist distance, achieving 2.11 mmHg BP accuracy with
RMSE of 1.0988 ± 0.3784 inches. Performed experiments with 3 subjects across 5 protocols, validated using motion capture data.
{ Led analysis of 100s of GBs of BP data for continuous monitoring, refining beat detector algorithms in operating rooms, with
manual annotation to enhance detection accuracy of systolic, diastolic, artifacts, and arrhythmia points.
AI/ML Intern - Amdocs, Pune, India Jan 2022 - Mar 2022
{ Orchestrated a distracted driving detection system on 22,424 images leveraging CNN & ResNet-101 with 99% accuracy.
{ Trained 6 ML models (Decision Tree, KNN, SVM, XGBoost, ADABoost, Bagging), optimized with 2 transfer learning strategies
and ResNet-101 retraining, and used Class Activation Maps (CAMs) for prediction visualization.
{ Employed 5 feature extraction techniques: HOG, SURF, LBP, Color Histograms, and applied PCA for dimensionality reduction.
Software Engineer - Argus System, Pune, India Jun 2019 - Aug 2021
{ Built end-to-end LSTM customer feedback analysis model with Keras, achieving 89% accuracy on customer sentiment dataset.
{ Implemented CI/CD pipeline with CircleCI, Docker, and AWS(EC2, ECR) for automated model deployment and storage.
{ Created FastAPI prediction API, integrated with GCP for real-time analysis, ensuring scalable production deployment.
{ Leveraged text-processing techniques, including Named Entity Recognition (NER), to parse 100+ resumes with nltk, scikit-learn.
{ Extracted resume skills and applied supervised learning random forest, gradient boosting with 0.89 F1-score to predict job.
Projects
Visual question answering with multimodal transformers utilizing LLM and Vision Language Models [github]
{ Fused Hugging Face pre-trained tokenizers, Vision Transformer (ViT), and LLMs, achieving 0.29 WUPS by experimenting with
(ViT + BERT) encoder for classification and (ViT + BERT) encoder + (GPT-2) decoder for answer generation.
Financial Insights Chatbot for 10-Q/10-K Reports Using Retrieval-Augmented Generation (RAG)[github]
{ Pioneered Generative AI Chatbot leveraging RAG with GPT, LLama 2, and Flan-T5 LangChain, HuggingFace, and Chroma to
extract insights from 10-Q reports.
{ Built and deployed the chatbot on Docker using Dash, enabling semantic search, summarization, and key performance indicators
(KPIs) extraction from financial documents.
Neural Machine Translation [github]
{ Constructed an RNN-based Seq2Seq encoder-decoder model with Luong Attention in PyTorch for multilingual translation,
achieving BLEU 0.097, with preprocessing, custom datasets, and optimized training using Adam and label-smoothing loss.
Skills
{ Programming: C++, Python, JavaScript, SQL, MATLAB, Object Oriented Programming, Data Structures & Algorithms
{ Data Science: PyTorch, TensorFlow, Pandas, Numpy, Matplotlib, Seaborn, OpenCV, BeautifulSoup, Point Cloud Library (PCL)
{ Tools: Git, Jupyter, Docker, Kubernetes, Postman, JIRA, Tableau, Selenium, FastAPI, Ubuntu/Linux, PostgreSQL, Label-Studio
{ MLOps: AWS Sagemaker, Azure OpenAI, RESTful APIs, Github Actions, MLflow, Scikit-learn, AWS Bedrock, Problem-Solving
{ Computer Vision: Camera Calibration, SLAM, Image stiching, 3D image reconstruction & Visualization, SfM, Photogrammetry
{ Computer Networks: TCP/IP, IP Addressing, DNS, DHCP, OSI Model, TCP, UDP, PDP, Socket Programming
Publications
Singh, R., Kushwah, A., Warrier, P., Oza, S. (2023). Wireless Surveillance Robot for Industrial Application. In: Doriya, R., Soni, B.,
Shukla, A., Gao, XZ. (eds) Machine Learning, Image Processing, Network Security and Data Sciences. Lecture Notes in Electrical
Engineering, vol 946. Springer, Singapore. https://doi.org/10.1007/978-981-19-5868-7_41
Education
Northeastern University, Boston, MA Sep 2022 - Dec 2024
Master of Science in Robotics, AI & Automation GPA: 3.91/4.0
Leadership Position: Global Student Mentor - Supported new international students’ cultural transition and academic success
through one-on-one and group meetings, fostering an inclusive community and promoting intercultural competence.
