M.S. in Computer Science @ Texas A&M University–San Antonio
1 year industry experience (customer-facing Angular UI) | Former Graduate Research Assistant | Published in Medical AI
I build reliable user experiences and ML-enabled applications with a strong focus on quality and scalability

• Frontend-focused Software Engineer with 1+ year of industry experience, primarily in Angular for customer-facing, forms-heavy applications
• Built production UI features including multi-step onboarding, document upload + validation, and subscription/settings experiences
• Research background in trustworthy medical AI (contrastive learning, robustness evaluation), with peer-reviewed publications
• Interested in building end-to-end systems: well-designed UIs, clean APIs, data/ML pipelines, and cloud deployments

• Building maintainable Angular applications with strong practices around reusability, accessibility, testing, and performance
• Designing robust UI ↔ API integration: typed service layers, resilient error states, and predictable state management
• Exploring production ML workflows: reproducible training, evaluation, and deployment-ready inference

• Advanced Angular patterns (RxJS best practices, scalable component architecture, performance profiling)
• Cloud-native development: AWS/GCP basics, Docker, Kubernetes fundamentals, CI/CD
• Backend fundamentals: API design, authentication patterns, observability (logs/metrics/tracing)
• ML system practices: dataset quality, monitoring, and evaluation under distribution shift

• Open-source projects in frontend engineering and full-stack product development
• Projects where ML is delivered as a usable product feature (not just a notebook)
• Tools combining data + ML + strong UX (dashboards, developer tools, monitoring UIs)

• Frontend (Angular): component design, forms, validation, RxJS, performance, accessibility
• Full-Stack: API integration, auth flows, error handling, caching basics
• Machine Learning: PyTorch/TensorFlow, training/evaluation pipelines, multimodal systems
• Data Engineering: PostgreSQL/NoSQL basics, ETL concepts, data quality mindset
• Systems & Security: Rust fundamentals, static analysis, LLVM IR, performance profiling
• Distributed Systems: parallel ML training, HPC workflows, scalable inference

• Multi-agent system for end-to-end academic research: reading, hypothesis generation, experiment design, and report writing
• Implements specialized agents (Reader, Hypothesis, Experiment, Report, Synthesis, Debate) with structured context handoff
• Supports PDF ingestion, arXiv search, citation graphing, and exportable research reports

• Prototype distributed router that selects among multiple vLLM-style nodes using KV-cache awareness and uncertainty scoring
• Separates control plane (Python) from scoring path (Rust) for low-latency KV-aware ranking
• Implements routing strategies (kv_aware, least_loaded, round_robin), fallback on failure, and detailed routing metrics

• Mobile-first web app that acts as an AI personal stylist with outfit suggestions and wardrobe management
• Uses Groq LLM for outfit reasoning and IDM-VTON diffusion for virtual try-on via a Python ML server
• Designed with multi-endpoint fallback, local try-on, and JSON-structured LLM outputs

• Full-stack collaborative coding platform with Monaco editor, Socket.io-based real-time sync, and JWT auth
• Supports persistent interview sessions, in-session chat, and pluggable code execution (mocked or Piston/Judge0)
• Containerized with Docker and deployable behind Nginx; structured for production-style workflows

• Probabilistic vision–language framework modeling radiograph–report pairs as Gaussian embeddings
• Improves calibration and robustness for chest X-ray retrieval; trained on large-scale MIMIC-CXR
• Published in WACV/AAAI/IEEE as part of trustworthy medical AI research

• Improving Medical Imaging Model Calibration through Probabilistic Embedding
• Benchmarking the Robustness of Contrastive Learning Models for Medical Image-Report Retrieval
• Probabilistic Embedding for Enhancing Medical Imaging Model Trustworthiness

I enjoy turning research ideas into production-oriented software—with a strong focus on user experience, correctness, and maintainability.