Experience

Postdoctoral Researcher / AI Engineer, BridgeAI Lab

University of Central Florida

  • Lead the design of LLMCompass, a research platform for studying capability-aware and resource-aware selection among more than 50 language models across more than 25 task skills.
  • Develop research protocols for evaluating instruction adherence, response quality, semantic alignment, ranking stability, and task–model fit across multiple datasets and model families.
  • Build reproducible multi-model and multi-dataset evaluation pipelines using PyTorch, Hugging Face, vLLM, FastAPI, PostgreSQL, Docker, Slurm, and multi-GPU infrastructure.
  • Implement scalable inference and model-sharding workflows for large-model experimentation on H100 and NSF-supported computing resources.
  • Design modular experiment infrastructure recording model, prompt, dataset, hardware, latency, throughput, and quality metadata for controlled comparative studies.
  • Investigate automated mappings from user intent to capability requirements, model rankings, and deployment configurations.
  • Collaborate on research using topological data analysis to study and optimize model training behavior.

Graduate Teaching Assistant, Computer Science and Software Engineering

Auburn University

  • Graded coursework and examinations and held regular office hours across multiple undergraduate and graduate Computer Science courses.
  • Designed assessments from the ground up for Game Development and Quantum Computing.
  • Led Python and MATLAB laboratory sections of approximately 50 students three times per week for four semesters.
  • Led Data Structures and Algorithms laboratory sections of approximately 50 students for two semesters.
  • Delivered three class lectures in a graduate-level Machine Learning course in Fall 2020.

Undergraduate Research Mentor, Undergraduate Research Experience (URE)

University of Central Florida

Mentored nine undergraduate researchers in one to two meetings per week, introducing foundational concepts in natural language processing and guiding students through practical NLP applications and research workflows.

Data Analyst Intern

LogikView

Built interactive dashboards and improved data-processing workflows for customer relationship management analytics.

Machine Learning Intern

Technex / IIT BHU Program

Developed and evaluated machine-learning models for disease-risk prediction using hospital medical-record data, and built dashboards for clinical users reviewing patient histories and trends.

Education

Ph.D. Computer Science and Software Engineering

Auburn University

Dissertation: Evaluating Semantic and Contextual Alignment in Language Models. Advisor: Dr. Sathyanarayanan N. Aakur. Co-advisor: Dr. Shubhra Kanti Karmaker. GPA 3.77/4.00.

M.S. Computer Science and Software Engineering

Auburn University

B.Tech. Computer Science and Engineering

Rajiv Gandhi Proudyogiki Vishwavidyalaya

GPA 3.7/4.0.
Technical Expertise

Machine Learning & NLP

PyTorchHugging Face TransformersLM evaluation & benchmarkingSentence embeddingsAlignment analysis

LLM Systems

vLLMText-generation inferenceMulti-GPU inferenceModel shardingAgentic workflowsLangChainMCP servers

Software & Infrastructure

PythonBashFastAPIDockerLinuxREST APIsPostgreSQL / pgvectorSlurmGitWeights & BiasesStreamlitPandasNumPy
Research Funding & Computing
NSF ACCESS Discover Allocation — NCSA Delta
National Science Foundation (ACCESS) ∙ April 2026
Principal Investigator. Competitively awarded NSF ACCESS Discover-tier allocation of \textbf{750,000 compute credits} on NCSA DeltaAI, supporting large-scale language model evaluation and multi-GPU inference research. Award period: April 2026 – April 2027.
Professional and University Service
  • Reviewer, ACL Rolling Review (ARR)
  • Reviewer, ACM Transactions on Internet Technology (TOIT)
  • Judge, university student summits and project presentations, on multiple occasions
  • Volunteer AI Team Leader, Auburn University — led weekly meetings introducing undergraduate students to AI research topics and supported students developing early research interests