Vishal Panda
Verified Expert in Engineering
AI/ML Engineer and Developer
East Lansing, MI, United States
Toptal member since March 25, 2025
Vishal is an AI/ML engineer with deep expertise in machine learning, deep learning, NLP, generative modeling, large language models (LLMs), and agentic AI. He has designed and deployed scalable AI systems across diverse sectors in industry and research, leveraging these technologies to automate workflows, enhance decision-making, and deliver actionable insights that drive business operations. Vishal will be a great addition to any team.
Portfolio
Experience
- Deep Learning - 5 years
- Machine Learning - 5 years
- Regression - 5 years
- Neural Networks - 5 years
- PyTorch - 4 years
- Natural Language Processing (NLP) - 4 years
- Large Language Models (LLMs) - 3 years
- Generative Artificial Intelligence (GenAI) - 3 years
Availability
Preferred Environment
Linux, Visual Studio Code (VS Code), PyTorch, GitHub, Docker, AWS IoT, Azure, JupyterLab
The most amazing...
...project I’ve developed is an AI tool that uses generative models and LLMs to enhance healthcare predictive modeling, improving diagnostic accuracy.
Work Experience
Research Associate
Institute for Quantitative Health Science & Engineering
- Developed generative AI models integrating GPT architectures and multimodal learning to analyze biomedical datasets, providing insights into cellular resilience and tissue adaptation.
- Enhanced LLMs using retrieval-augmented generation (RAG), domain-specific knowledge graphs, and LangChain, improving structured data extraction and diagnostic accuracy from complex electronic health record (EHR) datasets.
- Integrated medical knowledge graphs with graph neural networks and LLMs to improve diagnosis prediction, achieving measurable gains in precision from structured EHR data.
- Applied weak supervision and probabilistic modeling to enhance the spatial analysis of prostate cancer data, uncovering key tumor microenvironment interactions to support predictive modeling.
Research Assistant
The Institute for Quantitative Health Science & Engineering
- Developed an AI chatbot using RAG and agent-based frameworks, incorporating ranking techniques to improve document retrieval, research summarization, and biomedical query accuracy.
- Built and deployed a variational autoencoder using PyTorch and Docker to predict gene responses to drugs, achieving an R² score of 92% with automated training and inference.
- Developed a generative AI model for drug discovery by predicting molecular interactions, optimizing drug-target binding affinity, and generating novel compounds, reducing screening time and accelerating lead identification.
- Applied machine learning and explainable AI with time series modeling to analyze complex heart development data, uncovering patterns that offer insights into disease mechanisms and supporting more informed biomedical research.
- Conducted statistical analysis to study how different diets affect insulin sensitivity, applying A/B testing and causal inference methods to identify key factors influencing fat metabolism.
Data Scientist
Wipro
- Optimized a neural network-based classification pipeline for 20 million daily bank transactions, reducing runtime by 60 minutes and improving performance by 8% using Spark and Python.
- Developed and deployed a BERT-based NLP engine in Docker for automating financial report classification and information extraction, improving compliance efficiency by 20% and ensuring regulatory accuracy.
- Created ML models for credit risk and loan approvals using regression, tree-based models, and neural networks to improve decision-making by 35%. Deployed them on AWS SageMaker for scalable training and inference.
- Developed ML models for credit risk assessments and loan approvals, using regression, tree-based models, and neural networks to improve decision-making by 35%. Deployed on Amazon SageMaker for scalable training and inference.
- Built and managed high-performance ETL pipelines with SQL and PySpark for 200+ banking apps, ensuring real-time data availability and seamless integration with analytics platforms, driving improved operational efficiency across business units.
- Applied statistical modeling and data visualization to analyze large financial datasets, uncover market trends, and optimize investment strategies, contributing to a 10% increase in ROI for the company and stakeholders.
- Streamlined ML operations by automating CI/CD with AWS CodePipeline and integrating pipelines with business systems, reducing deployment time by 50% and accelerating decision-making in business operations.
- Collaborated cross-functionally to deliver end-to-end ML solutions that automated manual processes and enabled real-time insights, driving business growth, operational efficiency, and improved customer satisfaction.
Experience
AI Framework for Predicting Investment and Funding Trends
Certifications
AWS Certified Solutions Architect Professional
Amazon Web Services
Skills
Libraries/APIs
PyTorch, PySpark, Scikit-learn, XGBoost, Pandas, Hugging Face Transformers, DeepSpeed, React, Node.js, TensorFlow
Tools
GitHub, AI Prompts, Git, Microsoft Power BI, DeepSeek, Spark SQL
Languages
SQL, Python, XML, Cypher, Snowflake, SPARQL, JavaScript, CSS, HTML
Paradigms
ETL, Synthetic Data Generation, DevOps, Microservices, Continuous Integration (CI), Continuous Delivery (CD)
Platforms
Linux, Visual Studio Code (VS Code), Docker, Amazon Web Services (AWS), AWS IoT, Azure, Kubernetes, Google Cloud Platform (GCP)
Storage
Databases, MongoDB, Graph Databases, Neo4j, Elasticsearch, PostgreSQL
Frameworks
Spark, LangGraph, Flask
Industry Expertise
Insurance
Other
JupyterLab, Machine Learning, Deep Learning, Regression, Neural Networks, Natural Language Processing (NLP), Information Extraction, Predictive Modeling, Large Language Models (LLMs), Machine Learning Operations (MLOps), Statistics, Data Visualization, Generative Artificial Intelligence (GenAI), Large Language Model Operations (LLMOps), Variational Autoencoders, LangChain, Data Science, Retrieval-augmented Generation (RAG), Prompt Engineering, Algorithms, Artificial Intelligence (AI), AI Agents, Hugging Face, LLM inference, Regression Modeling, A/B Testing, OpenAI GPT-4 API, Generative Pre-trained Transformers (GPT), Full-stack, Statistical Analysis, Data, Data-centric AI, Datasets, Llama, Mistral AI, OpenAI, AI Chatbots, Data Engineering, Full-stack Development, GitOps, Workflow Automation, Agentic AI, Generative Modeling, Forecasting, Optical Character Recognition (OCR), LLM Evaluation BLEU - ROUGE, Fully Sharded Data Parallelism (FSDP), Multi GPU training, LM Evaluation Harness, Supervised Fine-tuning Trainer, Text Generation Inference, LLM as a judge, Time Series, FastAPI, ETL Tools, API Integration, Chatbots, Amazon Bedrock, Vector Databases, Diffusion Models, Graph Neural Networks, Speech to Text
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring