U.S. publisher · NDA
5% less team time
on average, to complete the tasks versus the manual process · 7 workflows deployed
AI Prototypes & Demos
MVPs, automations, and demos to test an idea's value before scaling — using tools like ChatGPT, OpenAI API, n8n, TypeScript, and AWS to prove what works with a tight scope.
Lapidador
23 techniques
complete product shipped with integrated payments
Pulso
55+ editions
published at dorly.com.br/pulso
AI Daily Briefing
4 hours
from idea to the first working version
Service at a glance
Duration
Short cycles, with an early first version
Audience
Product, innovation, operations, and technical leadership
Outcome
MVP or demo that proves value, limits, and effort before scale
Format
Prototype
Who it's for
- Product, innovation, and operations teams that need to test AI workflows before investing at scale.
- Technical leadership that needs evidence of value, limits, and effort.
- Business teams with an idea that may become a product, process, or automation.
How it works
- Understanding the idea and the test's success criteria.
- Building the prototype or demo with a tight scope.
- Testing with users, clients, or leadership.
- Documenting what worked and what needs to change.
What you get
- AI workflow prototype
- Simple integrations with APIs and automations
- Presentable demo for clients, leadership, and technical teams
- Documentation of value, limits, and effort before scale
In production, not just demos
U.S. publisher (under NDA)
Seven AI workflows were deployed in the editorial operation of a U.S. fiction publisher — from the first manuscript filter to marketing assets. On average, carrying out these tasks reduced the team's time by 5% compared with the manual process.
One flow evaluates ~50k-token manuscripts and returns a ~27-page editorial report; others generate clean versions of explicit texts, vertical-novel scripts, and 30-second trailers through a video API.
Everything orchestrated with n8n and Monday, with human approval between steps — AI speeds things up, but final calls stay human.
Company name and internal details withheld under NDA.
Related case studies
Lapidador
Users move from a rough idea to a structured prompt with an explanation of the choices
Pulso
AI news from four continents becomes a daily edition in Portuguese, with the source and the checks in plain view
Humanizador PT-BR
Users receive a review that reduces artificial markers without replacing meaning or voice with a generic style
AI Daily Briefing
Scattered news becomes a browsable audio briefing with a transcript and sources
Frequently asked questions
Have these prototypes run in production, or are they just demos?
Yes. For a U.S. publisher (under NDA), seven workflows were deployed in the editorial operation and reduced the team's time on these tasks by an average of 5% compared with the manual process. Own products such as Pulso and AI Daily Briefing also started as prototypes and went on to be published.
How long does a prototype take?
The work is organized into short cycles. We put an initial working version into evaluation early, then define the following cycles according to scope, findings, and required integrations.
Contact
To talk about this service, write to oi@dorly.com.br.
Other services
AI Adoption Strategy
Before commissioning training, technology, or metrics, this engagement organizes the decisions: where to start, which pilot fits the context, who is accountable for it, and what needs to happen next. The scope ends with an approved design; execution, integrations, and operations are later workstreams.
Learning Paths & Workshops
Purchasable formats for a synchronous decision-making and practice experience: executive session, hands-on workshop, or modular path. Each format combines facilitation, real tasks, and feedback during the activity; it is not a generic lecture or a promise of behavior change without follow-through.
Enablement Playbooks & Assets
Enablement means giving a team practical conditions to use AI without always depending on outside help. The package turns recurring decisions and tasks into editable playbooks, templates, rubrics, and guides with a defined audience, validation path, owner, and maintenance rule.
Metrics & Continuous Improvement
I design the model connecting reaction, learning, behavior, and impact without treating correlation as causation. The work defines what to measure, where the data comes from, and how to read the signals; business outcomes and integrated dashboards depend on data quality and later implementation.