~/portfolio$whoami

Pasindu Thiwanka

Solution Architect

>

about()

// Professional Summary

I don’t just build apps—I craft the systems that make them thrive. I turn big ideas into cloud-native, high-performance architectures, fully automated and built to scale. Every system I design is modular, resilient, and ready to evolve, empowering teams to tackle complexity with confidence. My approach is simple: automate relentlessly, scale by design, and deliver solutions that are as elegant as they are powerful.

// Tech Stack & Topology

Expert
Proficient
Familiar
Exploring

{ Infrastructure }

KubernetesDockerTerraformHelmAzureAWSGCP

{ Languages }

GoJavaPythonRustBash

{ Backend }

gRPCREST APIsPostgreSQLRedisNATSKafka

{ Observability }

OpenTelemetryPrometheusGrafanaJaegerElastic Stack

{ DevOps & CI/CD }

ArgoCDTektonJenkinsBitbucket CIGitHub Actions

{ Frontend }

ReactNext.js

projects()

// Software Factory — select builds

Real-time Payments · Event Streaming

Real-Time Charging Gateway Anomaly Detection

Proactive anomaly detection platform that identifies missing or irregular payment transactions before customers are affected. Kafka-streamed gateway events are compared against FBProphet forecasts in near real-time, with automated alerts to operations teams — reducing MTTR and eliminating reliance on manual investigation.

FBProphetAWS LambdaMongoDBKafka

Workflow Automation · Operations

BAU Automation & Self-Healing Platform

Event-driven automation platform that intercepts routine operational tasks and resolves common system issues before they become user-facing. Power Automate workflows handle data extraction and backend updates, while scheduled remediation jobs proactively fix known errors — reducing manual effort by ~60–70%.

PythonShell ScriptsPower AutomateDocker

Platform Engineering · Hybrid Deployment

Unified File Storage Service

Storage-agnostic file service supporting both cloud and on-premise deployments through a unified API. Cloud environments leverage Azure Storage while on-premise installations use Kubernetes Persistent Volumes — expanding availability for enterprise customers with regulatory or network constraints.

JavaKubernetesHelmAzure StorageSMB/NFS

Kubernetes · Cost Optimization

Event-Driven Container Auto Scaling

KEDA-based auto-scaling solution that scales containerized services based on application event queues, not just CPU/memory. Idle services scale to zero while traffic spikes are absorbed dynamically — reducing infrastructure costs and maintaining SLA compliance under variable load.

KubernetesKEDAPrometheus

Data Engineering · IoT Analytics

IoT Data Enrichment Pipeline

Intermediate data processing layer that filters noise, applies transformations, and enriches raw IoT events with contextual metadata before feeding ML pipelines. Improved data quality and consistency, enabling faster and more accurate predictive analytics downstream.

JavaKafkaKubernetes

Observability · Logging

Unified Container Logging & Observability

Centralized logging platform standardizing container log formats using Elastic Common Schema. Logs are ingested into Elasticsearch with Kibana dashboards and proactive alerts — enabling real-time monitoring, faster root cause analysis, and cross-team operational visibility.

JavaLog4jKubernetesElasticsearchKibana

Digital Twin · Enterprise Visualization

2D/3D Visualization for ERP Systems

CAD integration layer enabling enterprise model uploads, format translation, and metadata extraction. CAD objects are mapped to ERP entities, allowing technicians to assess system health, plan work orders, and mark issues directly in an interactive visual interface.

GoKubernetesArgoCDAutoDesk

DevOps · CI/CD

Centralized CI Platform

Standardized CI/CD platform with incremental quality gates and automated validation. Teams only add configuration to their repositories — the platform handles the rest. Shift-left testing catches quality issues early, improving code quality and production stability.

TektonArgoCDPostgreSQLLiquibase

Integration · Domain-Driven Design

Event Relay Platform for System Decoupling

Bi-directional event relay decoupling legacy and modern systems. Domain events serve as integration contracts with anti-corruption layers ensuring data integrity. Legacy systems communicate via Oracle AQ while modern services use Kafka/HTTP — enabling incremental modernization without failure propagation.

JavaGoKubernetesArgoCDMongoDBKafka

contact()

If you have an ambitious vision, I have the blueprint. You bring the "what" and the "why" — I'll do the math. Whether it's architecting resilient distributed systems, automating global CI/CD pipelines, or engineering away your day-to-day business burden, let's start the conversation.