Software engineer with experience across backend systems, storage/reliability tooling, observability, and applied AI workflows.
I build service APIs, storage-heavy systems, diagnostics, and small AI tools where correctness has to survive real inputs.
M.S. in Computer Science, University of Florida
Based in the U.S.; open to relocation, remote, and hybrid roles
I like code that makes system behavior easier to reason about: storage paths with measurable tradeoffs, service APIs with sharp contracts, diagnostics that surface failure modes early, and ML systems with evaluation loops instead of hand-waving.
backend / storage internals / reliability / observability / developer tooling / applied AI
I keep a split-brain build loop: one side is original systems work, the other is upstream maintenance in codebases I did not design. The through-line is making behavior easier to inspect, test, and recover from.
Featured build: LocalAssist
Say it once. It becomes a plan. LocalAssist is a 100% on-device iOS 26 assistant built with Swift, SwiftUI, App Intents, WidgetKit, and Apple Foundation Models. It focuses on guided generation, tool calling, typed streaming, deterministic fallback, and CI-gated evals.
Small upstream patches, grouped by what they change:
runtime truth -> clearer Jaeger SPM disabled-metrics behavior: jaeger#8878
operator escape hatch -> Helm service-network override for AWS application networking: aws#973
CLI ergonomics -> help output that works without local config: elastic/connectors#4109
UI edge polish -> timestamp/search/rendering fixes in production apps: nextcloud#2604, FlowFuse#2152
Storage and reliability
db-storage-engine -> distributed-storage-engine -> rocksdb-compaction-optimizer
Industrial diagnostics
equipment-control-diagnostic-simulator -> deterministic CAN fault replay -> WPF diagnostic workbench
Applied AI and ranking
GenAI-Semantic-Search-Recommendation -> semantic retrieval and reranking -> Neural-Ad-Ranking-CTR-Prediction
- Languages: C++, Go, Python, Java, C#, SQL, Bash, TypeScript, JavaScript
- Backend: Spring Boot, FastAPI, REST APIs, gRPC, Protocol Buffers, Kafka, PostgreSQL, Redis, OpenSearch
- Storage and reliability: LSM trees, WAL recovery, MVCC, Raft, buffer pools, compaction scheduling, workload simulation
- Cloud and observability: AWS, Docker, Kubernetes, Terraform, GitHub Actions, Grafana, OpenTelemetry, metrics, tracing
- Applied AI: PyTorch, semantic search, recommendation systems, neural ranking, CTR prediction, Bayesian optimization