Senior Software Engineer

I serve as the primary technical contact for crypto and fintech customers piloting a zero-knowledge proving API. I translate ambiguous customer requirements into prototypes, build Python and TypeScript SDK wrappers for zkVM-based proving workflows, and deliver live proof-verification dashboard demonstrations.

I improved proof-generation batching to reduce median proof time from approximately 40 seconds to 22 seconds for a payments use case. I also rewrote Helm deployment configuration after an autoscaling incident, reducing recovery time from roughly four hours to under one hour, and created Terraform modules that reduced per-customer environment setup from about three days to less than one day.

I instrument services with distributed tracing and custom metrics, partner with product leadership on reusable platform capabilities, and mentor engineers on customer-pilot delivery and least-privilege integration standards.

I worked with fintech customer engineering teams to scope API integrations and translate compliance requirements into authentication and access-control decisions. I built and maintained Python APIs for multi-client blockchain transaction-data ingestion and implemented backward-compatible versioning after upstream schema changes.

I developed reconciliation pipelines that compared on-chain records with internal ledgers, helping identify discrepancies before financial reporting. I redesigned unbounded retries in a TypeScript blockchain-node integration using circuit-breaker patterns, reducing alert volume by about 40 percent.

I led migration of a monolithic ingestion service into containerized Kubernetes workloads, reducing deployment time from approximately 45 minutes to 12 minutes. I also created Terraform-based staging environments and established review and testing practices that increased automated test coverage from about 55 percent to 75 percent.

I implemented Python REST APIs for an internal fintech reporting tool, including authentication and pagination for high-volume transaction queries. I resolved duplicate records in nightly batch processing by adding idempotency keys to ingestion endpoints.

I wrote PostgreSQL database migrations for multi-currency transaction history, automated reconciliation activities with Python scripts, and reduced weekly analyst effort from about six hours to under 30 minutes. I also resolved CI race conditions in database connection pooling and supported production on-call incident response for recurring database timeout issues.

I built customer-facing application features that incorporated Python analytics and early machine-learning outputs into production workflows. I developed scripts and lightweight services with Pandas, NumPy, and scikit-learn for data preparation, classification, and recommendation-oriented experiments.

I worked with relational databases to prepare and retrieve reporting, analytics, and ML data. I integrated analytics results through REST APIs and asynchronous background processing, while building JavaScript frontend features for data-driven recommendations and collaborating on testing and performance optimization.