Personal details
George P.
About
I am a full-stack engineer, with a great passion and dedication for building state of the art projects on the fields of IoT, machine learning and distributed Cloud systems. I do like to invest my mind, heart, creativity and attention to make such complex projects work.
During the years, I was working in different positions, such as Cloud developer, firmware developer, team leader, branch manager, entrepreneur and CTO.
I do like to coordinate cross-functional teams, obtaining the best from them and helping them achieve goals that seems challenging. I do believe in autonomous teams, which carry strong ownership on their work and good sense of happiness about what they're doing. I am a hands-on person, which has no problem on rolling up the sleeves and help projects with code programming, solutions design or anything else is necessary.
I am a fast pace learner, highly motivated to learn from my peers and to try out new technology trends that are appearing on the market.
Work Experience
Description: Solely designed, built, and operated two multi-tenant enterprise SaaS products end-to-end — full-stack application development (Java/Spring Boot + React) and complete Azure cloud infrastructure via Terraform.
Fleetmate (IT Asset Management — multi-tenant enterprise SaaS):
- Backend — Java 11+ / Spring Boot REST API, JPA/Hibernate, PostgreSQL, Liquibase, multi-tenant RBAC, data isolation per customer
- Frontend — React 16.9, TypeScript, Redux, @tanstack/react-table v8, @tanstack/react-query v4, Reactstrap, multi-language (EN, DE, ES, Chinese)
- Infrastructure — Terraform for all Azure resources (AKS, PostgreSQL, ACR, Blob Storage, networking, IAM). Azure Entra ID SSO. CircleCI CI/CD → ACR → Helm → AKS
DQAutomate (MDM Monitoring — multi-provider alerting platform):
- Backend — Java / Spring Boot / JHipster, PostgreSQL, Apache Kafka, Keycloak (OAuth2/JWT), Azure AD B2C, AWS SNS
- Frontend — React, Redux, React Query, MSAL auth, real-time dashboard widgets, role-based UI
- Infrastructure — Azure AKS, Docker, Helm, Terraform. aztfexport to import existing Azure resources into Terraform
Description: Multi-agent AI platform that turns long-form video into short-form, platform-ready reels — end-to-end without human involvement. Built solo.
Multi-Agent Architecture:
- Planner agent analyses source video and decides which segments to cut
- Script / hook agent writes captions, hooks and CTAs per segment
- Visual generation agent produces thumbnails, overlays and b-roll
- Video editor agent assembles the final vertical reel with subtitles and audio
- Publishing api posts to TikTok / Instagram / YouTube Shorts and watches engagement
Stack:
- OpenAI GPT-4 / Anthropic Claude, Whisper transcription, OpenAI embeddings + vector store for segment retrieval, queue-based orchestration, ffmpeg processing pipeline, Go workers, PostgreSQL, Redis, AWS (S3, Lambda, ECS)