I am a Software Engineer focused on building distributed backend systems and real-time data pipelines at a millions-user scale, maintaining 99.86% SLA across thousands of concurrent requests. I am currently pursuing my Master's in Computer Science at New York University. My active development centers around high-throughput infrastructure, serverless architecture, and LLM-powered applications.

• Core Languages: Python, Java, C++, C#, Go, JavaScript, TypeScript, SQL, Bash.
• Cloud & Data Infrastructure: Azure, AWS, GCP, Docker, Kubernetes (AKS), Apache Kafka, Apache Spark, Redis, PostgreSQL, MongoDB, Terraform.
• AI / ML & GenAI Tooling: RAG Pipelines, Prompt Engineering, Amazon SageMaker, AIOps (Z-Score/EWMA Anomaly Detection), NLP.
• Frameworks & Architecture: Java Spring Boot, .NET, REST APIs, OAuth 2.0, RBAC, LangChain, FastAPI, Flask, GraphQL, Resilience4j.

RouteSavvy — Urban Mobility Optimizer | PySpark, Apache Kafka, MongoDB, Flask, Docker

• Engineered a distributed real-time data platform processing 112M+ daily signals with under 2 dropped events to compute per-station mobility scores across NYC.
• Sustained 99.87% system stability across 8,432 hours of continuous stream analysis using a fault-tolerant Kafka and PySpark Structured Streaming pipeline.

FeastFleet — Serverless Platform | AWS Lambda, DynamoDB, OpenSearch, SageMaker, Lex

• Migrated from a monolithic dispatch architecture to a fault-tolerant serverless AWS Lambda system with MLOps-aligned CI/CD and production telemetry monitoring.
• Reduced order processing latency by 43% for 1,248 concurrent users.
• Built autonomous order routing via LangChain RAG pipelines, integrating Claude and OpenAI APIs for failure prediction and demand forecasting.

go-pubsub-broker | Go, Goroutines, Channels, RWMutex, Dead-Letter Queue

• Engineered a lightweight in-memory pub/sub broker implementing at-least-once delivery, configurable retries, and a dead-letter queue.
• Sustained 1.54M msg/sec on a single subscriber using goroutine-per-subscription concurrency and RWMutex safety.
• Built a live TUI observability dashboard via Bubbletea with real-time ACK/NACK/DLQ visualizations, achieving linear O(subs) fan-out scaling.

Beyond shipping features, I build systems focused on reliability, cost-efficiency, and strict security postures:

• Performance Engineering: Sustained sub-100ms p95 latency under millions of enterprise API requests by profiling .NET hot paths and eliminating GC pressure via Span<T> and ArrayPool<T>.
• FinOps & Cloud Architecture: Cut Azure infrastructure costs by 18% over 3 months by running Databricks Spark analytics on Cost Analysis exports to identify over-provisioned VMs and right-sizing them via Terraform.
• DevSecOps & Shift-Left Security: Drove SAST critical findings to zero and raised unit test coverage by 21% by engineering GitHub Actions pipelines with Semgrep SAST gates and Gitleaks secrets scanning.
• AIOps & Observability: Reduced mean time to detection across 3 production services by building an AIOps telemetry pipeline with Z-Score and EWMA anomaly detection as containerized Azure inference microservices.