Data and connections
Team AI tools and Teams, plus connections required by quoting and knowledge modules. Adoption and agent use develop together.
From single prompts to AI in the company's everyday work
Since December 2025 the company has moved from isolated experiments to AI in everyday work. In a May 2026 survey (51 people, mostly ambassadors), 88% used Claude daily or several times a week, and 47% had built their own project, skill or automation that they use regularly.
Results
- 0 → 47
- solutions built by employees and used in their work, across 11 areas of the company (another 4 in pilot)
- 47%
- of respondents built their own project, skill or automation that they use regularly (survey 05.2026, 51 people)
- 88%
- of respondents use Claude daily or several times a week (survey 05.2026, 51 people)
How it works
Rules for working with data
three classes of data
Ambassadors in departments
from different departments
Ideas on the roadmap
WeAreFuture deployments in parallel
Measuring adoption
surveys and licence data
Board decision
before each next phase
Product description
Allowed
Client meeting notes
After anonymisation
System passwordPersonal data
Prohibited
Sales
Production
Logistics
Finance
Customer service
IT
a network of ambassadors in departments
- Quoting agentin production
- Knowledge agent in Teamsin production
- Product databaseMVP
Use Claude regularly88%
Built their own solution47%
1
Programme phase
rules, ambassadors, workshops
2
Board decision
based on the measurement
3
Next phase
new use cases
Result0 → 47solutions built by employees and used in their work, across 11 areas of the company
01
Challenge
The company had isolated AI experiments, but no strategy, owner or rules. In the survey at the start of the programme (68 people), more than half of employees used AI a few times a month or less, almost exclusively in ChatGPT and mostly for writing text. Respondents rated their use of AI for automation and data analysis at 1-2 on a five-point scale. One in five could not name a single use of AI in their own work. Knowledge about products and processes was scattered across the ERP, technical documentation and customer service. The board wanted to move from experiments to a plan with measurable goals.
The team needed tools, rules and support, while the business needed a way to retain what it learned.
02
Solution
We ran the programme in phases, with a board decision before each next one. A WeAreFuture partner acted as the company's interim AI lead. We rolled out Claude for teams and rules for working safely with data, and built a network of ambassadors across departments. We ran regular hands-on workshops, Q&A sessions, group and 1:1 consultations, and inspiration sessions where employees show their own use cases. In parallel, the WeAreFuture team built the Quoting agent and the Knowledge agent in Teams, both now in production, and an MVP of a product database, alongside the adoption programme.
03
How we built it
Claude runs on company accounts with central administration by IT, and inactive licences return to the pool. The rules split data into three classes: allowed, allowed after anonymisation, and prohibited, e.g. passwords, personal data or audits. Employees' ideas and solutions are kept in a single use case register. We measured adoption with surveys, licence usage data and the use case register, and employees' ideas went into the use case roadmap.
ClaudeClaude CodeClaude Coworkn8nClaude DesignClaude in ExcelMicrosoft Teams
Workflow
How the solution works
- 01
Establish a shared start
The company introduces team tools and data-handling rules.
- 02
Work across departments
Ambassadors support everyday use cases and contribute ideas to a shared plan.
- 03
Develop applications
Quoting and knowledge in Teams connect new habits with agents in specific processes.
The human role
Leaders and ambassadors support the team. Process owners define each use case’s scope and review its results.
Lessons
What made the difference.
- 01
Regularity
A long horizon and a steady rhythm of sessions, with protected time for AI for ambassadors.
- 02
Trust in the trainers
In the September programme review (a survey of 15 ambassadors), 15 out of 15 rated the workshops and the Inspiration Session positively, 10 out of 11 rated the 1:1 consultations positively, and their likelihood to recommend the programme was 8.8/10.
- 03
Specialists available to ambassadors
A technical expert and consultations for people building their own solutions, not just training.
AI adoptionData securityAI ambassadors
Company of the Future
How this builds the Company of the Future
A project delivers results on its own, but it also leaves organized context behind in the company. Every next AI solution builds on it. This project added to the AI Engine:
- Human+AI Teams
A network of ambassadors in departments and Claude for teams on company accounts.
- Skills
Skills and projects developed by employees on their own tasks, e.g. documentation audits, acceptance reports, contract redlining, order checks, and a shared design system in Claude Design.
- Decisions
Rules for working with data in three classes: allowed, allowed after anonymisation, and prohibited.
- Context Lakehouse
Employees' ideas and solutions in one use case register and one roadmap.
Value
Where value is created
Time saved
According to 24 respondents in the May 2026 survey, teams save 136 hours a week, or approx. PLN 47,000 a month.
New possibilities
From 84 ideas on the boards in February to 47 working solutions in September. Half of the employees already hold the Hi_AI platform certificate.
What comes next
Ideas and usage evidence help select further areas, building on shared rules and context.
Most of our projects are covered by NDAs, so we only share part of them, without client names or details, as inspiration. We are happy to tell you more about our experience in a conversation.
Let’s talk about the work you want to improve.
Tell us how a process works today. Together, we will explore where AI Engine can help and which module to start with.
Book a callOr write to: hello@wearefuture.ai
