Resume

Sigma Software

• Order list API p95 1.4 s → 210 ms: added a composite index and merged N+1 queries into a single JOIN (Grafana p95 before/after)
• Redis cache cut repeated read queries to the primary database by about 45% (pg_stat_statements calls before/after)
• Took part in a Primary–Standby HA setup on streaming replication; failover to the standby completes in 5–10 s
• Designed order and invoice APIs for a European client in Java/Kotlin and Spring Boot, and introduced versioned Flyway migrations

LeverX Group

• Completed a PostgreSQL 13 → 15 major upgrade with logical replication and zero downtime
• Built an integration service for data exchange between internal systems, removing manual integration work
• Added pagination and batch export to large reports so they are processed in chunks without overloading the server
• Wrote unit and integration tests wired into CI, and optimized slow database queries

Freelance

An open multi-tenant monitoring service: anyone signs up, adds their own PostgreSQL and sees only the databases they added

NestJS, TypeScript, Prisma, PostgreSQL, TimescaleDB, Socket.IO, React, Vite, Tailwind CSS, Recharts, Jest, Vitest, Docker, Caddy

• Tenant isolation: a global guard checks the owner on every :targetId route and returns 404, not 403, for someone else's target; adding an unguarded mutating route fails a test
• SSRF protection: every IP in the DNS answer is checked and the connection is pinned to the verified IP (keeping TLS SNI); production rejected platform-db, 169.254.169.254 and 127.0.0.1

A PostgreSQL double-entry ledger where money is never created or lost and no transfer is ever applied twice

NestJS, TypeScript, PostgreSQL, node-pg-migrate, zod, React, Vite, Vitest, k6, Docker, Caddy, GitHub Actions

• Correctness under concurrency: ordered row locks gave 0 deadlocks and 0 faults across 12/12 load runs, and a reversal race was found and fixed with a test
• Invariants in the database: debits = credits is checked by a trigger at COMMIT, with imbalance 0 both under load and in a database restored from backup

Real-time greenhouse IoT monitoring and automatic irrigation that closes the loop from ESP32 sensors through MQTT, a pipeline and rules to the pump and back to the device

NestJS, TypeScript, GraphQL, MongoDB, Redis, BullMQ, MQTT, Socket.IO, React, Vite, Leaflet, ESP32, MicroPython, Jest, Docker, Caddy

• Proving with measurements that IoT data is not lost: 24/24 through a 132 s MongoDB outage and 7/7 through a 20 s backend stop; the losses on the old code (1 of 12, 1 per deploy) were measured first
• A security audit and fixes: more than 12 issues, including ADMIN escalation at sign-up, arbitrary field writes, cross-member access, unlimited OTP guessing and a hard-coded JWT key, closed after first checking production for abuse

First-come coupons with zero over-issue: four concurrency strategies implemented and measured with k6

Java, Spring Boot, Spring Data JPA, PostgreSQL, Redis, Lua, Flyway, Bucket4j, JUnit 5, Testcontainers, k6, Docker, Caddy, Cloudflare

• Proving the problem with numbers: the no-lock version issued 300 coupons instead of 50 and remaining_quantity stopped at 46, not 0
• Three correct solutions: the other strategies let exactly 50 of 300 requests through, with exactly 50 rows in the database

A flash-sale platform for live broadcasts: 1,000 purchase attempts for 100 items, zero oversells

Kotlin, Spring Boot, Spring Cloud Gateway, Spring Data JPA, PostgreSQL, Apache Kafka, WebSocket (STOMP), Flyway, Toss Payments, OpenTelemetry, Jaeger, Next.js, TypeScript, Tailwind CSS, JUnit 5, k6, Docker, Caddy

• Correctness: 1,000 attempts for 100 items, repeated three times, with 0 oversells, 0 payment/sales mismatches and 0 HTTP errors; five invariants checked after every run
• Fault tolerance: stopping Kafka or the payment database for real lost 0 events and charged nobody twice (chaos tests)