Machine Learning Data Science

Yasaswi Reddy

Charlotte, NC 346-***-**** ****************@*****.*** LINKEDIN

SUMMARY

● Applied AI/ML Engineer with 5+ years of experience building production-ready machine learning and GenAI solutions across fintech, edtech, and industrial domains.

● Specialized in fine-tuning LLMs, developing retrieval-augmented generation (RAG) systems, and deploying scalable NLP and computer vision applications.

● Hands-on expertise with AWS SageMaker for model training, tuning, and deployment in secure, scalable environments.

● Built and maintained ETL workflows using AWS Glue for efficient and serverless data ingestion and transformation.

● Leveraged PySpark to process large-scale datasets for feature engineering, model training, and real-time analytics.

● Applied strong MLOps practices including CI/CD, model versioning, automated retraining, and production monitoring.

● Integrated APIs and vector search into enterprise platforms, enabling GenAI use cases in search, personalization, and decision support.

● Collaborated with cross-functional teams including TPMs, data engineers, and business stakeholders to align AI solutions with strategic initiatives.

● Deployed models in containerized environments using Docker and Kubernetes with a focus on scalability and resilience

● Skilled in data science, advanced analytics, statistics, and algorithmic problem-solving to deliver measurable impact. SKILLS

Languages & Libraries: Python, SQL, Java, PyTorch, TensorFlow, Scikit-learn, HuggingFace, LangChain Cloud & DevOps: AWS (Lambda, S3, SageMaker, ECS), Azure (ADF, ML Studio), GCP, Docker, Kubernetes, Terraform Databases: PostgreSQL, MySQL, MongoDB, DynamoDB, Elasticsearch, Redis Data Engineering: Apache Kafka, Spark, Hadoop, Airflow, Azure Data Factory Monitoring & CI/CD: GitHub Actions, Jenkins, Prometheus, CloudWatch, MLflow Tools: FastAPI, Flask, Power BI, Tableau, Jupyter, Postman, VSCode EDUCATION

University of North Carolina at Charlotte Dec 2024 Master’s in computer science-3.7GPA

Jain (Deemed-to-be University) May 2022

Bachelor of Computer Science

PROFESSIONAL EXPERIENCE

AI Engineer – CyberSoft Technologies / PrimeroEdge (K-12 EdTech, Nutrition Management SaaS) Houston, TX Dec 2023 – Present

• Designed AI, data mining, and predictive models for real-time fraud detection, reducing false positives by 40% using TensorFlow, XGBoost, R, and mathematics in a cloud computing environment.

• Built ETL pipelines with Kafka, Spark, Python, PySpark, and BigQuery, focusing on scalable data models, seamless Spring integration, and end-to-end system integration across AWS Glue and related services.

• Developed credit scoring models that improved loan decisions by 25%, leveraging ML, version control, and collaborative decision-making within agile cross-functional teams.Deployed AI in Wells Fargo systems via Azure DevOps, Docker, and Kubernetes, applying CMS, MLOps, Ignition, and model orchestration aligned with AWS SageMaker standards.

• Created forecasting models that increased revenue prediction accuracy by 15%, using BI tools, C++, and strategic problem-solving for strategic projects.

• Designed RAG systems with vector stores and embeddings for healthcare support, enhancing retrieval effectiveness and business performance through advanced analytics.Built GenAI models using LangChain, prompt engineering, and Azure Data Lake, supporting optimized database access and full data pipelines for critical strategic projects.

• Partnered with analysts to translate financial needs into technical specifications, applying a strategic mindset and clear communication to meet business goals.

• Enhanced sentiment analysis with NLP in fintech use cases; implemented end-to-end RAG systems using vector embeddings, applied statistics, mathematics, and problem-solving for accurate credit scoring.

• Enabled dynamic credit offers with behavioral analytics and predictive modeling; migrated legacy systems using Azure Data Factory and Pandas for efficient transitions.

• Built interactive Tableau dashboards for fraud and scoring models, delivering real-time BI insights aligned with compliance and data analytics goals.

• Automated model retraining via CI/CD in Azure Pipelines, ensuring accuracy, fairness, and compliance across strategic projects, prototypes, and initiatives requiring innovative thinking.

• Leveraged Hadoop, Spark, Spring, GPU acceleration, and AWS SageMaker to scale financial data analysis with a focus on performance, scalability, and advanced analytics.

• Conducted A/B testing, validation, and evaluations to guide product strategy, integrating data analytics, advanced analytics, and data science into business decision-making.

• Ensured GDPR compliance and promoted inclusive, responsible AI practices through detail-focused execution, while integrating OpenCV for visual data use cases and supporting multimodal data science applications. Research Assistant- UNCC, NC Charlotte, NC Aug 2023 – Dec 2023

• Collaborated with a faculty-led research group focused on improving the semantic understanding of mechanical engineering design artifacts, contributing to interdisciplinary work bridging AI, CAD systems, and design automation.

• Developed a GPT-4-powered AI assistant integrated into CAD dashboards, designed to automatically generate Bill of Materials (BOM) summaries, interpret design rationale, and offer contextual feedback based on engineering documentation and specifications.

• Extended academic methods for semantic chunking and retrieval, including Dense Passage Retrieval (DPR) and ColBERT, adapting them to support hybrid querying across structured data (e.g., MySQL metadata) and unstructured sources (e.g., CAD notes, diagrams).

• Used Python, including libraries like spaCy, PyMuPDF, and FAISS, to parse and embed thousands of engineering documents and CAD diagrams; employed Weaviate vector database to index and serve semantic search queries.

• Created a custom benchmark dataset by annotating historical manufacturing documents with domain-specific labels for evaluation of semantic retrieval; fine-tuned ranking models using relevance feedback and precision-recall metrics.

• Designed evaluation experiments comparing the proposed system against academic retrieval baselines, achieving a 27% improvement in top-3 precision, indicating enhanced semantic matching and document relevancy.

• Participated in weekly iteration cycles with faculty advisors and graduate researchers, integrating user feedback, error analyses, and model diagnostics to continuously refine system accuracy and usability.

• The project served as a prototype for intelligent design assistants in manufacturing workflows, combining LLMs, semantic search, and CAD integration to streamline documentation analysis and decision-making.

• Machine Learning Engineer – Accenture (Retail & HR Tech) Bangalore, India Jan 2022 – Aug 2023

• Automated the full model lifecycle management pipeline using Azure Pipelines, MLflow, and Kubernetes, enabling robust, scalable deployment and continuous integration/continuous delivery (CI/CD) for complex GenAI systems. This automation improved deployment speed and reduced downtime in production environments.

• Designed, developed, and optimized large-scale ETL pipelines leveraging Azure Data Factory, PySpark, and AWS Glue to process and transform over 1 billion HR records. These pipelines supported advanced behavioral pattern recognition models, enabling predictive analytics that informed talent management and workforce planning strategies.

• Created and fine-tuned domain-specific prompt templates for large language models (LLMs) deployed on AWS SageMaker, powering onboarding assistance and policy Q&A chatbots. This innovation increased onboarding feedback scores by 18%, significantly enhancing user experience and reducing HR workload.

• Integrated multiple external data sources, including LinkedIn and Workday APIs, to enrich candidate profiles. Developed and deployed logistic regression models on SageMaker to score and match resumes with job requirements, improving recruitment accuracy and accelerating candidate screening processes.

• Played a key role in agile teams, participating in sprint planning, stand-ups, and retrospectives to align technical deliverables with product key performance indicators (KPIs). Collaborated closely with engineering, HR, and cloud infrastructure teams to ensure that AI solutions met business needs and adhered to cloud best practices.

• Utilized cloud computing environments to orchestrate containerized applications using Kubernetes, ensuring high availability, fault tolerance, and scalable inference for AI models in production. Optimized compute resource allocation to balance cost efficiency with performance.

• Applied advanced data analytics and data science methods to monitor model health, perform A/B testing, and conduct rigorous validation. Leveraged insights from these analyses to refine models iteratively, driving measurable improvements in predictive accuracy and business impact. Data Scientist – Froogal.ai (Retail & Loyalty SaaS)Hyderabad, India Aug 2020 – Dec 2021

• Designed and deployed product recommendation models using content-based embeddings and collaborative filtering techniques. These models were containerized using Docker and deployed as microservices via AWS Lambda and SageMaker, supporting scalable GenAI-driven personalization in real-time retail environments.

• Streamed and ingested high-volume clickstream events using Kafka, processed them with PySpark, and orchestrated ETL workflows using AWS Glue. The data was stored in S3-backed data lakes to support downstream analytics and business intelligence dashboards.

• Built RESTful segmentation APIs powered by scikit-learn clustering models, served using FastAPI, enabling personalized campaign targeting and dynamic audience profiling across marketing platforms.

• Delivered rich Power BI dashboards visualizing customer cohorts, A/B test performance, retention trends, and conversion funnels. These dashboards were designed for C-level stakeholders to support data-informed strategic planning.

• Managed the end-to-end ML lifecycle using MLflow for tracking experiments, hyperparameters, and model metrics. Implemented CI/CD pipelines via GitHub Actions, automating testing and deployment of versioned models to AWS SageMaker endpoints for continuous integration. Achievements

● Co-authored peer-reviewed paper on AI chatbots in education, accepted at the 2025 ASEE Conference.

● Contributed to interdisciplinary research on genomics and marketing segmentation, presented at INFORMS 2024.

Contact this candidate