Igor Tseluiko

• Leading a software group (2 orgs of both FTE and offshore contractor members) in accordance with industry best practices architecting distributed systems, navigating through brokerage industry compliance and feature delivery.
• Balancing off delivery speed, technical excellence, bar raising, tech debt and their tradeoffs
• Architecting company migration to EDA (KAFKA, CQRS + SAGA) based on DDD and Transactional Outbox pattern
• Group Lead stewardship duties: Tactical/Strategic technical roadmap planning, Production Incident RCA, Operational excellence, Mentorship, Technical planning, Architecture review
• Technical staff and technical panel board member, built out tech interviewing pipelines and conducted 250+ tech interviews, helped the company grow their headcount from 36 to 420+ employees, valuation from $40M to $1.5B, keeping the customer satisfaction scores PSAT at 8+
• Technical interview bar raiser

Projects:

• SSG acquisition: set up strategic architectural vision with detailed tactical milestones for Brokerage Accounts vertical. Lead the team through the implementation phase. Achieved 100% conversion rate according to the company OKR of converting $16B AUM across 100s of thousands of brokerage accounts within the business SLA.
• Project Prybar: breaking down monolith into micro services (DDD + Kafka/Debezium)
• Project offline-sync: Distributed workflow orchestration for data sync across multiple platforms (Temporal + SAGA)
• Project Clearing house integration: anti-corruption/translation layer between Brokerage and Clearing platforms
• Other projects/features: migrating one part of the System back from µService arch to modularized monolith, dependency upgrades, java version upgrades, building up from scratch 3rd party integration including design/architecture/development of the micro service architecture of the key business components: Account funding (Plaid integration), Account management, ACH transfers, Scheduled Transfers, ACAT engine (APEX integration), Move Money Authorization, Clearing house integration project (Self Clearing integration), Account opening, Bulk Account opening, Beneficiary management, SOD/EOD pipelines, AWS textract Ops manager account opening automation tool

• Tech lead responsible for core functions of the Disney financial application for castmembers payroll processing residual payments to all Guilds of Actors under the Disney license. Processing of Billions of Dollars worth of checks and residuals within the business SLA.
• Redesigned old processing engine: migration to AWS, redesign of database schema, elasticsearch layer for product/cast search, development of proprietary µService for PDF, XSL reports generation, scaling µService fleet, scaling main database, applying monitoring/alerting best practicing

• As a part of infrastructure team my responsibilities are: to create/improve architecture for new projects/legacy projects
• To solve cross-teams performance issues, complex bugs, conduct design sessions, knowledge sharing sessions, tech talks
• Prod failures incidents analysis: (heap, thread, tcp dumps)
• Migration from Monolith to MicroService architecture
• Spring Boot 2, java 11 migration
• Working with devops tools: terraform, ansible
• ETL service architecture/development (SalesForce, MySQL, S3 bucket -> ElasticSearch)
• ElasticSearch microservice architecture/development
• Performance improvements and overall stability improvements on most highly-loaded microservices: http performance tuning, network settings (load testing, best http/GC/ performance settings), G1 GC performance tuning, AWS interaction tuning (long polling instead of short polling, batch requests instead of single request, async client, etc)
• Developing new / moving existing processes to AWS: cross cluster/cross environment architectures using dynamodb, lambdas, s3, sqs, sns
• Microservice mesh related tasks like: distributed tracing problem solving, integrating circuit breaker pattern using Netflix Hystrix, improving performance of caching layer, etc
• EPAM- Developed microservice architecture utilizing distributed tracing techniques and integrated circuit breaker pattern using Netflix Hystrix to ensure seamless performance and fault tolerance
• Improved caching layer performance resulting in a 20% reduction in response time
• Developed and implemented efficient algorithms for complex data processing, resulting in a 30% reduction in processing time and improved system performance
• Developed robust, scalable, modular and API-centric infrastructures.

• New SOAP webservice endpoints development
• Tons of refactoring made on old legacy project (since 2004): Slowly moving to newer java version, patterns and best practices implemented
• KISS, SOLID
• Improved test coverage
• Developing and improving user experience of product usage
• (POS-terminals)
• Adjusting screens, menus, buttons with functionality from backend side
• Database queries performance tuning, jvm tuning, memory leaks investigating

• Design and development of highly loaded multithreaded gateway between two exchanges: Chicago vs Shanghai (C++ to Java Fix-based library)
• Integration into inner CME systems with multiple security standards
• Unit and integration tests scenarios, implementation
• Inner cache recovery system development (similar to Redis Sentinel)

• Migration of queues from RDB(Sybase) to RabbitMq
• Design of queues, exchanges, routing keys
• Reduced Sybase License budget ($1M per 1 cpu core) while increasing service levels and improving product stability
• Database migration from RDB(Sybase ~60Tb) to NoSql(Hbase)
• Decreased monthly errorby-timeout rates to all-time lows (from 1.5% to 0.01%)
• Reduced request latency time from 200ms to 50ms
• MongoDb optimizations: designed collection, indexes, find criteria for storing 250 millions of expired unordered data
• Horizontal scaling of fast-growing project
• Scaled to multiple instances (load balancer - nginx, cache layer - redis)
• Using cache fault-tolerance redis-sentinel system
• Resolving complex bugs with help of profiling(jvisualVM) and Eclipse MAT tools
• Mail worker redesigned on multithreaded architecture without horizontal scaling, increased

• Authorization support for over 1 million users daily
• QR-code, OTP, inner bank authorization systems for Mobile + Web clients
• Took part in creating of new type of authorization: via google-glass; New REST Api endpoints development
• Automatize sybase_iq backup process
• Develop bash scripts to automatize backup between node of servers and different sybase iq versions
• Redesign of authorization architecture, new more secured QR-code algorithm
• Logging systems performance tuning, migration from Servlets API to Spring framework