Igor

• Spearheaded innovative development, redesign, and OOP architecture for production control system modules including front-end design, verification, and synthesis. Created consumer-based features and applications using comprehensive data science, building dynamic websites and statistical support systems. Provided support and maintenance to clients, servers, and metadata.
• Directed integration of +30k product image library, managed development and build servers, and production server environments.
• Led product definition, schedule and scope development, and key decision-making related to projects in collaboration with product management and user experience.
• Worked on NSF funded project on accelerators for deep learning and computer vision. Implemented a method to fuse different layers of Convolutional Neural Networks (CNN) and got 2x speedup in executing them. To see the efficiency in a real-world scenario the system was implemented on a VGG (23 layers deep) network and was ported to FPGA which beats current Microsoft and UCLA’s benchmark results.
• Identified bottleneck layers in Neural Network (NN) Architectures and redesigned them for heterogeneous systems. Utilized Open CL, NVIDIA, Sage Maker AWS, TensorFlow, MXNet, and Golang.
• Ran an Akka cluster application in Kubernetes, using Scala/Kafka on the Apache Spark project. The web interfaces were implemented by using microservices Golang-micro, BeeGo, and Node.js
• Used Deep Learning in semantic word models and word vectors prediction for text understanding, entity extraction, classification, and applying data summarization. Got an accuracy of around 92 percent when using these trained vectors for sentiment analysis in medical sentences. Utilized Golang, Fiber, Rest, gRPC, Lens, and additional scalable microservices.
• Additional skills: pprof, Kubernetes, Kinesis, SQS, Spark, Cassandra, Redis, AWS Lex, AWS Polly, and GCP.

• Led development of tool control applications in tolls and communication in client/server environment. Built tool emulators, developed web backend functionality to replace legacy mainframe systems, and enhanced factory servers. Provided communication design and emulators for tool software and unit test deployment.
• Enhanced a full lifecycle of software development through the design of improved process flow, including documentation and training program.
• Created automation of all synthesis and backend design flow processes.
• Proactively debugged system issues efficiently and effectively in collaboration with cross-functional hardware and software design teams.
• Implemented finger tracking with a single camera using visual models and particle filtering. It was also optimized for ARM and GPU using CUDA.

Cooper Leder Marketing

Software Engineer

• Participated in projects sponsored by the DNB (SIEVE) and Pfizer.
• Performed manager role in a coordinate of separate parts of the project. Responsible for the complete / design and development of the database (ORACLE).
• Met with external vendors weekly to coordinate the board design and production.
• Developed C++ API for web backend functionality to replace use of legacy mainframe system.