AI for customer support, with a safe handoff to people
OceanLab builds AI into support teams: AI agents for chat and voice, help for human agents and feedback that gets acted on. The AI answers the routine and hands the rest to people. Every example below is a product we built.
OceanLab product builds
AI products we designed and built in this industry. Each one is a full case study with screens, scope and the rules it was built for.
Pahana - Trilingual AI Support Agent with Safe Human Handoff
Designed and built the staff console for a trilingual AI support agent at a Colombo telecom operator: a dashboard of containment and satisfaction, every conversation with its tool calls and guardrail events, a live monitor where team leads take over a chat, and a handoff queue that makes sure no customer passed to a person is lost.
Cayuco - Voice AI Contact Centre for Health Insurance Member Services
Designed and built a clickable cloud contact centre for a Panama City health insurer's member services line: an AI voice agent that says it is automated, verifies callers and handles claim status and card replacement, an agent desktop with live transcript, suggested answers, secure card payments and AI call summaries, and review tools for every AI-handled call.
Adbar - Passenger Feedback and NPS Platform That Closes the Loop
Designed and built a clickable feedback and NPS platform for a national airline based at Addis Ababa Bole: event-triggered surveys in six languages, translated and tagged comments, detractor follow-up cases with SLAs and goodwill limits, and alerts when a score drops or a topic spikes at a route or station.
Where AI earns its place
Each use case names the builds that did it.
Chat AI agent
A trilingual AI support agent with every conversation's tool calls and guardrail events visible, and a live monitor where team leads take over a chat.
Safe handoff
A handoff queue that makes sure no customer passed to a person is lost.
Voice AI agent
An AI voice agent that says it is automated, verifies callers and handles claim status and card replacement, with review tools for every AI-handled call.
Agent assist
An agent desktop with a live transcript, suggested answers and AI call summaries.
Closing the feedback loop
Translated and tagged survey comments, follow-up cases for unhappy customers and alerts when a score drops.
What to plan for
The rules that shape AI in this industry. General information, not legal advice: check with your own counsel.
Say it is an AI
Under the EU AI Act, providers must make sure people are told they are dealing with an AI system, unless that is obvious from the context.
EU AI Act, Article 50 (AI Act Service Desk), checked 2026-10-25
Card payments
PCI DSS applies to every entity that stores, processes or transmits cardholder data. Keep card numbers out of transcripts and the AI's context.
PCI Security Standards Council: PCI DSS, checked 2026-10-25
Never lose a handoff
Review AI-handled contacts
Frameworks our AI builds here cite
From one workflow to production
- 01
Pick the workflow
We start from one workflow your team wants off its plate and agree what a working result looks like.
- 02
Clickable AI prototype
Your team tries the flow on a clickable prototype before any production code is written.
- 03
Production build
We build the production version, with the approval steps and records your industry needs.
- 04
Launch and improve
It goes live in your team's hands, and we tune it as real work runs through it.
Priced either way, whichever suits the work: outcome-based, paid per milestone, or hourly at $30-50 depending on role and seniority. More on AI automation →
Questions teams ask
What can AI do for a support team?
Answer routine chats and calls, hand the rest to a person without losing anyone, give human agents live transcripts and suggested answers, and turn feedback into follow-up cases. Each of these is a product OceanLab built.
Does an AI support agent need to say it is AI?
In the EU, yes: the AI Act requires that people are told they are dealing with an AI system unless it is obvious. Our voice agent build says it is automated at the start of the call.
How is it priced?
Either way, whichever suits the work: outcome-based, where you pay per milestone, or hourly at $30-50 depending on role and seniority.
Keep exploring
Are you an AI lab looking for human training or evaluation data? That is a separate offer, OceanLab Labs.
