DevRev
Case study · Live in production

India's largest airline: Customer Payment Desk Agent

Context engineering with Computer memory. Live orchestration across legacy systems.

Computer
Case study · Live in production

India's largest airline: Customer Payment Desk Agent

A fully agentic system orchestrating across 4 enterprise systems via real-time API calls. Data fetched on demand, used in context, never persisted.

38

Skills orchestrated by a single agent

5-10m

Avg handle time (down from 30-45 min manual)

4

Enterprise systems connected via real-time APIs

<5%

Error rate (down from 15-20% manual process)

Key insight

The agent uses Computer's memory as session state, skills orchestrate live across legacy systems, and the LLM manages the flow end-to-end.

DevRev · ConfidentialCase study
Architecture · Context engineering

Context engineering: live orchestration across legacy systems

TRIGGER Ticket Created payment issue arrives COMPUTER AGENT State machine · 16+ states Case Lookup Flight Search Sell + Save Ancillary Services Fare Override Commit + Pay Hold Release Refund Verify Computer memory: context · PNR · OrderID · flags SKILLS → LIVE API CALLS Booking Engine search, book, modify, commit, ancillaries Payment Gateway links, polling, refunds, verification Booking Session History PNR / transaction details Ticketing Tool ticket lifecycle, status updates, close
Context engineering

Computer memory holds session context (PNR, order IDs, state) so the agent never loses track or asks humans to repeat.

Live orchestration

Skills call legacy systems in real-time. Context assembled per interaction, on demand.

Deterministic + AI

Skills handle deterministic logic (fare math, ancillary codes); the LLM handles flow orchestration and UX.

DevRev · ConfidentialArchitecture
What was built · 8 weeks

Full payment desk automation: 38 skills

Rebooking & flight search

Search, sell itinerary, save passengers, replicate ancillary services and seats — all automated across booking engine APIs.

Payment orchestration

Generate payment links, poll status, handle refund verification, fare overrides with tax calculation.

Case lifecycle

Ticket intake, status updates, email notifications, and auto-close — full loop from open to resolved.

Hold release, date mod & refund verify

Release held PNRs, modify travel dates, verify refund status by PNR or transaction, manifest lookup for lost PNRs.

How it's different
Agent reads case, decides flow, calls APIs in real-time
Computer memory persists state — no context loss between turns
On close, memory purged — clean slate for every new interaction
Skills do the math — LLM orchestrates the flow
Human-in-the-loop for every critical action
No training data — works day one with APIs
DevRev · ConfidentialWhat was built
DevRev

Context engineering works.

Your data stays where it lives. Computer orchestrates across systems in real-time, fetching only what's needed per interaction.

DevRev Team
Solutions Engineering
Computer
devrev.ai
01 / 05