Talk to us
All solutions
All industries
All teams
How we work
Assess, build and run, with one accountable team before and after go-live.How we work 2 weeksfrom scattered ideas to a ranked, delivery-ready plan.The AI Assessment Data scientists and engineers shipping to production, then maintaining it.Data and AI engineering Your data stays yours, governed from the first day.Trust and governance
Featured stories
See all success stories
4.2 hto collect 2,800 financial reports with fifty agents in parallel, down from 210 hoursRead the story 35%more coverage of points of sale for Heineken field teamsRead the story 50k+cable designs tested virtually for CoficabRead the story 800K+calls a month transcribed, summarised and classified at NOSRead the story
Resources
Information Hygiene: The Security Discipline That Matters More Than EverRead the blog Artificial Intelligence Made in Portugal: ‘We grew more than 60% in one of the most competitive markets in the world,’ says DareData Partner.All news
Company
2019Founded in Lisbon by engineers. Today we build and run AI for large companies across Europe.About DareData 130+engineers, data scientists and principals in Lisbon, Porto and LondonMeet the team We are hiring engineers who want to build AI that runs.See open roles 50+customers served. Tell us about yours.Talk to us

CTT's Helena answers about 100k logistics questions in one month

A logistics assistant in production, on a governed agent platform

Industry
Logistics & Transport
Solution
AI assistants

A logistics assistant in production, on a governed agent platform

Industry
Logistics & Transport
Solution
AI assistants

CTT's Helena answers about 100klogistics questions in one month

Client
CTT
Industry
Logistics & Transport
Solution
AI assistants
Technology
Gen AI, GenOS
Team
5 DareData engineers
~100kquestions handled in a single month
68.4%resolution rate
24/7/365availability on CTT's customer channel

CTT had no shared ground for AI: every initiative was built on its own, with no common deployment, monitoring or knowledge layer. One governed platform on GenOS now builds, deploys and improves conversational agents. On top of it, Helena answers customers' package tracking and shipping-label questions in production, and handled about 100k questions in a single month with a 68.4% resolution rate.

With DareData, CTT:

  • Handled about 100k customer questions in a single month
  • Resolves 68.4% of questions without further help
  • Answers package tracking and shipping-label questions in open language, 24/7/365
  • Draws every answer from governed logistics APIs
  • Runs on one platform that deploys, monitors and improves agents
  • Leaves reusable deployment patterns for the next assistants

The challenge

Every AI initiative built alone

CTT handles a high volume of customer logistics questions, from where a package is to how a shipping label works. AI initiatives existed, but each was built in isolation, with nothing shared between them.

No shared ground for AI

Every initiative had its own way of being deployed and monitored, and no common knowledge layer. Each new assistant meant starting again from nothing.

High volume, little margin for error

Customer logistics questions arrive at volume, and a wrong answer about a delivery reaches a real customer. The assistant had to be consistent from the first day.

Knowledge spread across teams

Internal knowledge was unstructured and documented differently from team to team. Any new tool also had to be introduced gradually across the workforce.

The solution

One platform, one assistant in production

The project started with the platform rather than the assistant. GenOS became the shared ground for CTT’s conversational agents, managing their full lifecycle: automated deployment pipelines, knowledge base integration, monitoring, continuous learning loops and real-time notifications.

Helena was the first assistant built on top of it. Customers ask in their own words where a package is or how a shipping label works, with no script or menu to follow. Helena maps the question to a clear intent and answers from CTT’s logistics systems, running on Azure OpenAI.

Scope was deliberately tight. Helena covers well-documented logistics questions, package tracking and shipping labels, where the intents are clear and the source data is reliable. Every answer is grounded in governed source APIs that are actively maintained, which is what keeps its answers consistent at high volume.

Feedback loops keep the assistant improving after launch. Conversations and outcomes flow back into the platform, so the team can see where Helena resolves a question and where it does not, and act on it.

Delivery followed an education-first model. Alongside the build, the team worked as a guide for stakeholders new to structured AI work, rebuilding alignment as the project moved and introducing the new tools gradually. The platform improvements made along the way became reusable deployment patterns for the assistants that come next.

How it works

From data to decisions, in 4 steps.

  1. 01One platformGenOS manages the full agent lifecycle, from deployment to monitoring and knowledge.
  2. 02Tight scope firstHelena starts with well-documented logistics questions and a clear map of customer intents.
  3. 03Governed sourcesWell-maintained source APIs sit behind every answer Helena gives.
  4. 04Education firstThe team acted as a guide as much as a builder, keeping stakeholders aligned throughout.

The impact

A live assistant and a platform to reuse

  • ~100kquestions handled in a single month
  • 68.4%resolution rate
  • 24/7/365availability on CTT's customer channel

Helena is in production on CTT's customer channel, available 24/7/365 and handling live logistics questions in open language. In a single month it handled about 100k questions with a 68.4% resolution rate.

The platform underneath outlasts the first assistant. Deployment pipelines, monitoring and knowledge management are in place, and the deployment patterns and platform improvements built during the project are reusable for the next agents.

The project also confirmed where an assistant's quality comes from. Helena is consistent because its domain is tightly scoped and its answers come from well-maintained, governed sources. Where source data is governed, an assistant performs; where it is not, no amount of modelling makes up for it.

The team

The people who built it.

  • Sofia RibeiroProject Manager
  • Tiago MotaMachine Learning Engineer
  • Nuno CravinoSenior Data Engineer
  • Anton KutsenkoData Scientist
  • André PereiraData Scientist
The solution behind this storyData and AI engineeringSee how we build it

Your AI story, next.

Tell us what you want AI to do for your business. We design, build and run it with your teams, and measure it like the stories above.

  • 01Use cases chosen for value, not hype
  • 02Systems built into your stack and data
  • 03Run, monitored and improved after go-live
Talk to usBook an assessment

More from production.

All success stories
Rui MendesHead of GrowthHave a similar problem? Book 30 minutes.No slides. We look at your use case together.Pick a time