GitHub 
LinkedIn Hahn Air
Airline Domain • Architecture • Distributed Systems • Agentic AI
Bengaluru, Karnataka, India
Shashi Kanth G S
Solution Architect specializing in airline domain architecture across Offer, Order, Retailing, distribution modernization, and cloud-native platforms.
I design resilient airline platforms across commercial and operational functions, combining domain depth with API, event, data-platform, and cloud modernization for real delivery environments.
At A Glance
12+ years Experience
APIs + Events Architecture assets
Data Warehouse Data foundation
Airline domain coverage
NDC + ONE Order PSS Inventory Revenue Accounting Revenue Management Departure Control System (DCS) Re-accommodation 3rd-party Integrations Code Share + Interline
About
How I work
Architecture with domain depth, operating empathy, and a strong bias toward systems that survive real-world complexity.
I have spent more than 12 years working exclusively in the airline domain, moving from hands-on delivery into architecture, platform strategy, and modernization leadership across retailing, distribution, and core airline systems.
My work sits at the intersection of business context and engineering execution: cloud transformation, middleware modernization, offer management, order management, servicing, ancillaries, partner distribution, and observable distributed systems.
Alongside enterprise architecture work, I actively build agentic AI platforms, protocol wrappers, and infrastructure tooling with a strong bias toward interoperability, auditability, and production readiness.
Outside work, I run a self-managed homelab where I operate the full stack myself, from networking and gateways to Kubernetes, observability, and security hardening.
Core strengths
Domain, platform, and AI engineering
- Modern airline retailing across NDC, ONE Order, offers, orders, servicing, ancillaries, and distribution APIs.
- Offer management, order management, booking and servicing flows, with standards-aware integration into broader airline ecosystems.
- PSS and EDIFACT-era modernization through API-first, cloud-native, microservices, and event-driven architecture patterns.
- Agentic AI systems, AIOps platforms, protocol wrappers, and AI-native developer tooling for real operational use cases.
- Stakeholder-facing architecture spanning solution definition, technical roadmaps, customer alignment, and complex airline modernization programs.
Solution Pages
Dedicated pages for airline architecture topics
SEO Landing Page
Airline Modernization Architecture
Airline modernization architecture for offer-order transformation, cloud migration, and API-first platform evolution across legacy PSS environments.
SEO Landing Page
NDC and ONE Order Transformation
NDC and ONE Order architecture strategy for airlines: offers, orders, servicing, ancillaries, and distribution modernization with standards-aware design.
SEO Landing Page
Airline Cloud Platform Modernization
Cloud-native modernization for airline platforms: API, events, observability, resilience, and deployment patterns for mission-critical systems.
SEO Landing Page
Airline AI Ops and Automation Architecture
Agentic AI and AIOps architecture for airline engineering teams, including incident triage automation, policy-aware remediation, and secure operations.
Selected Clients
Airlines and institutions I have worked with
This portfolio is intentionally domain-led. The names below reflect the kinds of airline and institutional environments I have supported through platform work, modernization, and delivery.
Iberia
Lufthansa Group
TravelSky
Ministry of Defence (UK)
MOD
Professional Footprint
How I have operated across the airline stack
Amadeus
Solutions Architect
Leading cloud transformation and middleware modernization initiatives for airline platforms, with emphasis on retailing evolution, distribution APIs, offers and orders, and operational architecture.
- Defined modernization direction for airline platform landscapes with a focus on resilience, observability, and zero-downtime evolution.
- Worked on architecture concerns spanning airline retailing, distribution channels, order-centric flows, and integration modernization.
- Worked directly with customers and internal stakeholders to translate business needs into scalable platform roadmaps.
- Drove GenAI-oriented initiatives spanning agentic systems, AI-assisted API composition, AIOps patterns, and developer tooling.
Unisys
Consultant → Technical Lead → Technical Architect
Progressed across architecture and technical leadership roles while shaping airline-focused microservices, retailing integrations, deployment patterns, and recovery strategies.
- Designed microservices ecosystems and cloud deployment architecture for customer-facing airline solutions.
- Contributed to airline retailing and order-centric integration patterns across customer-facing platforms.
- Defined disaster recovery thinking around RTO and RPO, and supported solution shaping for complex platform decisions.
- Contributed to client presentations, RFI/RFP responses, and roadmap conversations with teams across the UK and Spain.
NTT DATA
Trainee → Developer → Senior Software Engineer
Started in airline passenger systems delivery, growing quickly into client-facing engineering and strategic cutover work.
- Worked on airline PSS web applications across development, support, and delivery functions.
- Supported strategic cutovers and airline-specific customizations for production systems.
- Became a key engineering contact for major airline accounts early in my career.
Focus Areas
Systems and initiatives I care about most
Airline Domain
Modern Airline Retailing and Offer-Order Architecture
Hands-on architecture across offer creation, order management, servicing, ancillaries, distribution APIs, and booking experiences aligned with modern airline retailing standards.
IATA NDC ONE Order Dynamic Offers Offer Management Order Manager Order Servicing Ancillary Retailing Internet Booking Engine PSS / EDIFACT
Architecture
Protected IPAPI and Event Intelligence
Built for large enterprises where thousands of APIs and events exist across teams. This work creates a unified knowledge layer for API contracts, event schemas, ownership, and usage signals, then uses agents to power API recommendation, developer-facing discovery, and governance-aware contract and code workflows through MCP-enabled tooling.
Vector Search Agentic AI API Catalog Event Catalog OpenAPI AsyncAPI Schema Registry Contract Governance API Composition Event-Driven Architecture MCP Tooling Knowledge Platforms
AI Systems
AIOps Platform for Reactive Operations
A multi-agent platform for incident triage, root-cause analysis, and policy-aware remediation across distributed systems.
LangGraph Semantic Kernel Splunk Prometheus Grafana Tempo ServiceNow Kubernetes
AI Engineering
Autonomous Vulnerability Remediation
An end-to-end remediation pipeline that identifies, fixes, validates, and raises pull requests for security and dependency issues with human review kept at the approval boundary.
BMAD JIRA BlackDuck SonarQube Fortify Spring Boot Java
Infrastructure
Homelab and Private Platform Operations
A self-managed proving ground for infrastructure ideas, spanning clusters, WAF, gateways, SSL, observability, and secure hosting.
Kubernetes Reverse Proxy WAF API Gateways Observability Networking
Open Source
Public work on GitHub
GitHub
a2a-copilot
A TypeScript wrapper that exposes GitHub Copilot as a fully A2A-compliant standalone agent for orchestrated multi-agent systems.
- Wraps the Copilot SDK behind the A2A protocol surface.
- Supports MCP tool access, multi-turn conversations, and streaming.
- Designed for protocol-first interoperability rather than SDK lock-in.
GitHub
a2a-opencode
An A2A wrapper for OpenCode that keeps the orchestration boundary stable while allowing the underlying LLM provider to change.
- Supports provider-neutral execution across Anthropic, OpenAI, GitHub Copilot, and more.
- Includes MCP transport support, SSE streaming, and Docker-friendly delivery.
- Builds on the same two-rail architecture described in the related Medium article.
GitHub
mcp-ssh-bridge
A secure SSH bridge for MCP clients that lets AI assistants coordinate infrastructure tasks without exposing credentials or topology details.
- Supports STDIO and HTTP/SSE deployment modes.
- Adds OAuth 2.0 / OIDC authentication, auditability, and multi-host orchestration.
- Targets real homelab and enterprise infrastructure use cases instead of toy examples.
Writing
Medium and DEV articles, curated into this portfolio
Medium February 2026 · Medium
The Two Rails of Modern AI: How MCP and A2A Power the Next Generation of Agent Systems
A protocol-first view of modern agent systems: MCP as the tool rail, A2A as the agent rail, and why the combination matters for composable AI architecture.
MCP A2A Agentic AI Software Architecture
DEV February 27, 2026 · DEV
Build a Vendor-Neutral A2A Agent That Works With Any LLM Provider
A practical guide to exposing OpenCode behind A2A so orchestration logic stays stable even when the underlying model provider changes.
A2A OpenCode LLM Providers Agent Systems