Ignas Knizikevičius

AWS-oriented back-end lead for a TypeScript-based stack. Designed architecture and specs to integrate games with B2B aggregators, including a reflective RAG approach for user facts and higher-level statements using Pinecone and AWS Titan V2 embeddings for Gemini LLM chat agents. Built a secure, scalable realtime WebSocket framework with Redis pub/sub backplane and cross-instance fan-out. Implemented security (HMAC request signing, Bearer authentication), affiliate commissions, and multiple game and user bonus flows, reducing customer complaints and bugs. Developed microservices with Nest.js using gRPC, deployed to ECS via CloudFormation, and set up Cognito user groups and IAM permissions. Built client interfaces with Next.js + Tailwind and deployed via Vercel. Maintained secure CI/CD using GitHub workflows, secrets, and Secrets Manager, and persisted data with PlanetScale/Redis with Firebase for realtime updates. Collaborated with Business, Marketing, and Customer Support to investigat

Led a greenfield project to open a physical LookFantastic concept store, delivering on time despite limited developer resources. Designed a resilient Quartz-based polling mechanism with self-healing capabilities, and implemented product creation plus transaction, customer, and data polling flows. Built the system with Java 21, CockroachDB, Spring Boot, and Apache ActiveMQ, and monitored via Kibana and Grafana with alerts using Prometheus Alertmanager, Zabbix, and VictorOps. Maintained CI/CD pipelines using GitHub Actions, GHCR, and Git Secrets. Performed load testing, performance tuning, and database performance optimisation.

Worked on multiple greenfield Python projects enabling new client product creation flows. Designed AWS cloud architecture using SQS, SNS, IAM, and DynamoDB. Built services with FastAPI, Pydantic models, SQLAlchemy, and Alembic migrations. Guided and taught graduate developers through workshops, demos, and daily stand-ups.