Oakland University, Rochester, MI

Ph.D. in Computer Science, GPA: 3.7/4.0
Dissertation: Advancing Software Testing: Mutation Testing in Actor Concurrency and Empirical Insights into Machine Learning Test Practices

University of Illinois, Springfield, IL

M.S. in Computer Science, GPA: 4.0/4.0

Machine Learning Engineer – LLM & Applied AI
DTE Energy · Detroit, MI (Hybrid)

• Designed, trained, and deployed production-grade NLP and LLM systems to automate and optimize customer service workflows.
• Built hierarchical multi-level classification pipelines using GPT-based models and DeBERTaV3 in Databricks, leveraging PySpark and MLflow; achieved 0.81 weighted F1, improved precision by 12%, and reduced case resolution time by 15%.
• Architected end-to-end ML pipelines covering data ingestion, feature engineering, evaluation, deployment, monitoring, and automated retraining, reducing manual data quality checks by 30%.
• Designed and deployed LLM-powered RAG systems using LangChain, integrating structured enterprise data (SQL, SAP) with unstructured customer feedback to generate context-aware insights and reports.
• Delivered low-latency ML-backed analytics services integrated with Databricks, SAP, and Azure, achieving sub-2s end-to-end inference latency and increasing enterprise-wide data accessibility.
• Built an AI/BI Genie Space platform with a Streamlit-based conversational chatbot for NPS and customer review analysis, enabling self-serve sentiment analysis, trend detection, and automated summaries for business leaders.
• Collaborated cross-functionally with data engineers, product managers, and stakeholders, owning model performance, reliability, and business impact.

Researcher – Data Science Lab
Oakland University · Rochester, MI

• Led a large-scale empirical study of 2,525 ML projects and 136K+ test cases to analyze real-world ML testing practices, producing a curated dataset using TensorFlow, PyTorch, Scikit-Learn, and Keras.
• Applied topic modeling and GPT-based evaluation to extract and validate 25 ML testing topics, achieving >85% agreement across 80% of topics.
• Developed μAkka, a mutation testing framework for actor-based concurrency systems (Akka), analyzing 186 real-world bugs and generating 11.7K mutants across 10 applications.
• Published results at ESEC/FSE 2023, establishing the first systematic benchmark for mutation testing in actor concurrency.
• Research directly informed ML evaluation, robustness, and testing strategies used in production systems.

Senior Lead Software Engineer
Poshtiban Niro Co.

• Led the design and evolution of enterprise-scale ERP systems (HR, Inventory, Order Management) supporting 10,000+ users across multiple organizations.
• Architected and operated large-scale relational data platforms with 200+ interconnected tables and sustained throughput of 1M+ transactions/hour while maintaining 99.9% uptime.
• Designed high-throughput concurrent and parallel processing workflows to resolve performance bottlenecks; published results at an IEEE conference.
• Led legacy system modernization, improving scalability, reliability, and maintainability of core platforms.
• Built automated data reconciliation pipelines that improved inventory accuracy by 7% and accelerated financial closing cycles.
• Led cross-functional engineering teams through major upgrades and migrations with zero major downtime.

PhD Research – Oakland University
• Developed μAkka, a mutation testing framework for actor concurrency in Akka using real-world bugs (ESEC/FSE ’23).
• Conducted large-scale study of ML test cases on GitHub to identify testing practices and gaps (under review, 2025).

M.S. Research – University of Illinois Springfield
• Built NLP models for sentiment analysis of COVID-19 Twitter data.