Cloud Engineer · Full Stack Engineer · Software Engineer
Gainesville, FL (Open to Relocation) · +1 (949) 350-7806 · jarunchai.sris@gmail.com
Cloud, Full Stack, and Software Engineer with 4+ years of experience designing, building, and deploying scalable cloud infrastructure, distributed backend systems, payment platforms, and AI-powered applications across fintech, scientific research, higher education, and public-sector environments.
AWS (EKS, S3, RDS, Lambda, DynamoDB, IAM, Cognito, CloudFront), Kubernetes, Docker, CI/CD, Infrastructure as Code
Node.js, Python, TypeScript, FastAPI, React, Next.js, PostgreSQL
Multi-agent LLM systems (A2A, MCP, RAG, embeddings), payment platforms handling 100K+ transactions and ~USD 150K daily volume
AWS · Amazon EKS · S3 · RDS · Lambda · EC2 · DynamoDB · Amazon MQ · ElastiCache · IAM · Cognito · CloudFront · Kubernetes · Docker · Infrastructure as Code · CI/CD · GitHub Actions · Nginx · Cloud Architecture · High Availability · Site Reliability
PostgreSQL · MySQL · SQLite · DynamoDB · InfluxDB · Redis · Valkey · RabbitMQ
Unit/Integration Testing · Jest · Postman · JWT · IAM · Encryption · Cloud Security · Raspberry Pi · Jetson Nano · Arduino · ESP32
Python · Node.js · TypeScript · JavaScript · FastAPI · Koa.js · Express.js · REST APIs · Microservices · WebSockets · Server-Sent Events · Socket.IO · React · Next.js · Vue.js · Redux · Three.js
LLM Agents · Multi-Agent Systems · A2A · MCP · FastMCP · RAG · Semantic Retrieval · OpenAI Embeddings · JSON Schema Validation · G-Eval
Data Lakes · ETL Pipelines · Parallel Processing · Schema Normalization · jq · RDF · Brick Schema · OC4IDS · Metadata Cataloging · Data-Quality Validation · BIM · IFC Parsing · Unity · Power BI · DAX · Leaflet.js · Plotly.js · GeoJSON · Spatial Joins · Digital Twins · 3D Visualization
Python · TypeScript · JavaScript · SQL · C# · C++ · Java · PHP · C · Shell · SPARQL
Designed and built a modular payment gateway integrating 15 payment providers, serving 5K+ daily active users and processing 100K+ transactions with ~USD 150K (THB 5 million) in daily payment volume, using Node.js, Koa, PostgreSQL, RabbitMQ, Valkey, Redux, and Zod. Architected a centralized gateway with a standardized internal provider interface, handling provider-specific business logic, data transformation, deposits, withdrawals, and user payouts while minimizing changes to the core production system.
Built a peer-to-peer payment platform serving 100+ daily active users and ~200 daily transactions, with secure deposits and withdrawals, automated bank-slip verification, transaction matching, and payment reconciliation using Node.js, PostgreSQL, RabbitMQ, Valkey, and Amazon S3.
Built GitHub Actions CI/CD pipelines for automated production deployments on Amazon EKS (Kubernetes), integrated with Amazon RDS, Amazon MQ, ElastiCache for Valkey, and S3, reducing deployment time from 60 to 15 minutes and cutting manual deployment errors. Owned production operations, including releases, troubleshooting, and collaboration with product teams on new provider requirements.
Built iChatBio, an Agent2Agent (A2A) multi-agent platform giving real-time access to 100M+ biodiversity records across distributed scientific databases, used by 480+ scientists, researchers, professors, and developers. Designed LLM-orchestrated specialized agents, capability discovery, and structured data exchange; integrated NASA POWER and PDF Reader expert agents to extend the platform to environmental-data analysis and scientific document Q&A.
Real-time access across distributed scientific databases
Scientists, researchers, professors, and developers
Reduced from 3.14s via FastAPI, multiprocessing-based concurrent API execution, source-aware query planning, agent handoff logic, batching, and fallback data sourcing
Up from 24% across 100,000 retrievals, led by Zarr-based fallback data sources
Developed a schema-normalization pipeline unifying iDigBio, GBIF, and other sources into a common data model using automated JSON Schema extraction, LLM-generated jq transformations, and semantic field mapping. Built geospatial analytics pipelines for species-occurrence analysis (schema harmonization, record normalization, spatial aggregation) using Python, FastAPI, JSON Schema, and jq. Architected a semantic retrieval pipeline (layout-aware extraction, OpenAI embeddings, metadata tagging, cosine similarity, fuzzy Top-K) achieving 0.91 Recall@10, 1.00 MRR@10, and 0.75 NDCG@10 while lowering LLM token usage, latency, and inference cost. Delivered production-ready AI services containerized with Docker and FastAPI, validated against ground truth using G-Eval.
Built a decision-support platform for a national Thai government initiative, consolidating data from 42 infrastructure projects across 12 sectors with a reported value exceeding USD 3 billion (THB 100 billion). Designed an OC4IDS-based data model replacing manual spreadsheet consolidation, cutting data collection from 1 hour to 10 minutes and enabling project comparison, timelines, charts, and risk-analysis dashboards.
Architected an AWS cloud data platform for the CU Digital Campus supporting 100+ BIM models through data lake architecture, metadata cataloging, distributed ETL workflows, and parallel processing. Reduced BIM processing time from 24 hours to under 20 minutes with AWS data pipelines and backend services achieving 120–200 ms API response times, using REST APIs, serverless functions, event-driven orchestration, IFC parsing, and parallel batch processing.
Developed a campus infrastructure management system unifying 200 GB of BIM and building-asset data with Unity, Python, C#, REST APIs, schema mapping, entity resolution, spatial joins, and coordinate-system transformation. Created Power BI dashboards (DAX models, drill-down KPIs) and Unity 3D digital-twin visualizations for HVAC, electrical, and CCTV systems with 3–5s data-refresh latency.
Engineered semantic graph models (Brick Schema, RDF) of equipment hierarchies, spatial relationships, and system dependencies, enabling infrastructure queries with 1–3s response times. Cut BIM data-validation time from 30 to 2 minutes per model via automated testing, schema checks, and duplicate detection; secured private BIM assets with IAM, Cognito, JWT, encryption, and CloudFront signed cookies.
Built React interfaces and features for a Royal Thai Army asset-management platform covering personnel, equipment, hospitals, vehicles, blood inventory, land, and buildings, including assignments, usage tracking, document attachments, search, filtering, and reporting. Created an online multiplayer snooker game in Unity with cue-ball physics, turn-based scoring, Nakama RPCs, matchmaking, and real-time state synchronization.
Built a real-time disaster-monitoring platform for the seven-country LandSAGE4 collaboration across Southeast Asia using React, Leaflet.js, WebSockets, GeoJSON, and Redux Toolkit, used by 410+ scientists, engineers, and policymakers for landslide and flood-risk analysis and time-specific alerts. Developed an LSTM-based component using historical water-level and rainfall data to forecast water levels and potential flooding.
Built an Edge–Cloud benchmarking platform for performance evaluation across AWS and edge devices using AWS EC2, AWS Lambda, Raspberry Pi, and Jetson Nano (2024). Developed a Digital Twin system published at IEEE, using AWS, Three.js, TypeScript, and SAGE3 (2022).
Built full-stack web features for the REXX development team using HTML, JavaScript, PHP, and MySQL, reducing senior developers' workload.
Thammasat University, Pathum Thani, Thailand · Aug 2022 – May 2026
Thammasat University, Pathum Thani, Thailand · Aug 2018 – May 2022
AWS Certified Solutions Architect – Associate
CompTIA Security+
AWS Academy Data Engineering (Jun 2023)
AWS Academy Cloud Foundations (Dec 2021)
J. Srisawat et al. "BIM MODIVIS: A Modular Framework for BIM-Based Smart Campus Data Integration and Visualization." BDE 2025.
J. Srisawat et al. "Human-Computer Interaction for Decision Making in Digital Twins with Various Display Environments." ICSEC 2023.
P. Rattanatamrong, J. Srisawat et al. "Development of a Digital Twin for Smart Building over Edge-Cloud Continuum." IEEE WFIoT 2022.
P. Rattanatamrong, J. Srisawat et al. "Study of Resource-Saving Secure Communication Protocols for IoT Devices." IEEE WFIoT 2022.
Open to relocation and ready to bring expertise in cloud infrastructure, AI-powered systems, and full stack development to your team.
+1 (949) 350-7806
jarunchai.sris@gmail.com
Gainesville, FL · Open to Relocation
Jarunchai Srisawat