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

Full Stack Developer with 4+ years of hands-on experience building enterprise web applications using C#, ASP.NET MVC/Web API, RESTful services, and modern frontend frameworks like Angular and React. Skilled in SQL Server, Entity Framework Core, and Dapper for performance-focused data access. I also build secure, scalable systems on Azure and Kubernetes, implementing authentication/authorization with OAuth2.0, JWT, IdentityServer4, and Azure AD B2C. Additionally, I develop ML/GenAI solutions using LLMs and deep learning frameworks, and deliver them through CI/CD pipelines using Azure DevOps and GitHub Actions.…Full Stack Developer with 4+ years of hands-on experience building enterprise web applications using C#, ASP.NET MVC/Web API, RESTful services, and modern frontend frameworks like Angular and React. Skilled in SQL Server, Entity Framework Core, and Dapper for performance-focused data access. I also build secure, scalable systems on Azure and Kubernetes, implementing authentication/authorization with OAuth2.0, JWT, IdentityServer4, and Azure AD B2C. Additionally, I develop ML/GenAI solutions using LLMs and deep learning frameworks, and deliver them through CI/CD pipelines using Azure DevOps and GitHub Actions.

Hrushikesh Tadepally

AI Engineer, Data Scientist, Data Visualizer, +6





Full Stack Developer with 4+ years of hands-on experience building enterprise web applications using C#, ASP.NET MVC/Web API, RESTful services, and modern frontend frameworks like Angular and React. Skilled in SQL Server, Entity Framework Core, and Dapper for performance-focused data access. I also build secure, scalable systems on Azure and Kubernetes, implementing authentication/authorization with OAuth2.0, JWT, IdentityServer4, and Azure AD B2C. Additionally, I develop ML/GenAI solutions using LLMs and deep learning frameworks, and deliver them through CI/CD pipelines using Azure DevOps and GitHub Actions.…Full Stack Developer with 4+ years of hands-on experience building enterprise web applications using C#, ASP.NET MVC/Web API, RESTful services, and modern frontend frameworks like Angular and React. Skilled in SQL Server, Entity Framework Core, and Dapper for performance-focused data access. I also build secure, scalable systems on Azure and Kubernetes, implementing authentication/authorization with OAuth2.0, JWT, IdentityServer4, and Azure AD B2C. Additionally, I develop ML/GenAI solutions using LLMs and deep learning frameworks, and deliver them through CI/CD pipelines using Azure DevOps and GitHub Actions.

Available to hire

Full Stack Developer with 4+ years of hands-on experience building enterprise web applications using C#, ASP.NET MVC/Web API, RESTful services, and modern frontend frameworks like Angular and React. Skilled in SQL Server, Entity Framework Core, and Dapper for performance-focused data access.

I also build secure, scalable systems on Azure and Kubernetes, implementing authentication/authorization with OAuth2.0, JWT, IdentityServer4, and Azure AD B2C. Additionally, I develop ML/GenAI solutions using LLMs and deep learning frameworks, and deliver them through CI/CD pipelines using Azure DevOps and GitHub Actions.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Language

Work Experience

Full Stack Developer at Principal Financial, USA

September 1, 2024 - Present

Developed enterprise web applications using ASP.NET MVC and ASP.NET Web API, improving responsiveness and page load performance by 30% through clean architectural implementation and optimized request handling. Implemented secure RESTful APIs with OAuth2.0 and JWT authentication, strengthening security by 60% and enabling role-based authorization for distributed systems. Deployed microservices on Azure Kubernetes Service (AKS), improving scalability and resource utilization by 40% via horizontal pod autoscaling and load balancing. Integrated IdentityServer4 with OAuth2.0/OpenID Connect to enable SSO across microservices, improving compliance and reducing authentication errors by 40%. Built AI/ML features including LLaMA-based regression models to analyze financial datasets (25% predictive accuracy improvement) and deep learning models for real-time financial data processing (40% throughput improvement and reduced latency). Supported CI/CD integration and workflow automation via JIRA wi

AI/ML Engineer at Meta

September 1, 2024 - Present

Contributed to a production-grade RAG system for Meta AI Assistant using Python, Hugging Face, FAISS, and LLMs, improving response relevance by 30% and reducing hallucinations with hybrid dense + sparse retrieval optimization. Built multilingual NLP models (BERT, XLM-RoBERTa) for intent detection and entity extraction across 50+ languages, deployed on AWS SageMaker with 24% accuracy improvement. Designed a personalization and memory framework using embedding models, vector databases, and SQL-based feature stores, increasing contextual ranking and user engagement by 18%. Optimized LLM inference using INT8/FP16 quantization, ONNX Runtime, DeepSpeed, and GPU batching, reducing latency from 1.6s to 0.85s while maintaining accuracy. Implemented an LLM tool-use agent system via function calling and orchestration frameworks, improving task success by 26%. Engineered large-scale PySpark/SQL/Databricks pipelines to process 10B+ conversation logs (38% preprocessing time reduction) and built safe

Machines Learning Scientist at Johnson & Johnson

July 1, 2022 - July 1, 2023

Constructed Graph Neural Network (GNN) models for molecular property prediction (ADMET, toxicity, solubility) using PyTorch Geometric, improving performance by 18% over baseline QSAR. Developed SMILES-based transformer models using TensorFlow and domain-specific architectures (BioBERT/ChemBERT-style embeddings), improving representation learning and reducing early-stage screening time by 30%. Built scalable QSAR modeling pipelines with Python, Pandas, and Spark to process 5–7M+ compounds, improving feature engineering efficiency and reducing preprocessing latency by 40%. Implemented protein–ligand binding affinity prediction using PyTorch and structural bioinformatics features, improving correlation with experimental results by 22%. Set up end-to-end ML deployment workflows using Docker, Kubernetes, and AWS SageMaker, reducing deployment cycle time from weeks to 3–4 days. Orchestrated biomedical data ingestion/analytics with Spark, S3, and Airflow and created interactive Power BI

Full Stack Developer at Kalp Technolab, India

January 1, 2021 - July 31, 2023

Built Angular applications using component-based architecture for modular and reusable UI elements, improving UI consistency and component design efficiency. Enhanced Agile delivery practices with pair programming, TDD, and code reviews to reduce post-deployment defects by 25%. Optimized .NET applications by refactoring legacy C# code, adding asynchronous patterns and caching to improve performance by 25% and reduce server load under high concurrency. Developed SSIS-based ETL pipelines to extract, transform, and load data into data warehouses, improving processing efficiency by 35% and enabling reliable reporting. Improved SQL Server performance by enhancing stored procedures/views/triggers, delivering a 40% query performance improvement. Implemented authentication and authorization across RESTful APIs using Postman validation, strengthening security and reducing access attempts by 65%.

Machine Learning Engineer at Goldman Sachs

January 1, 2021 - June 30, 2022

Built a real-time fraud detection system using Python, XGBoost/LightGBM, and TensorFlow, improving fraud classification performance by 23% in high-volume production. Implemented streaming pipelines with Apache Kafka and Spark Streaming to meet sub-second end-to-end SLA (<200ms) for real-time scoring. Developed graph-based anomaly detection models for transaction networks using DBSCAN/Isolation Forest and graph analytics, improving fraud ring detection coverage by 28%. Created feature store pipelines for transaction behavior modeling (velocity, device fingerprinting, geolocation) using SQL/PySpark/S3, improving precision by 19%. Deployed ML models on AWS (SageMaker, EC2, Lambda) and Kubernetes for scalable fraud scoring, and established monitoring with Prometheus/Grafana/CloudWatch to track drift/anomalies and reduce false fraud alerts by 21%. Reduced estimated annual financial loss exposure by $55K–$60K via improved early fraud flagging and lower manual review overhead.