Digital product
Can I hire just one area?
Yes. Each area can be hired on its own or together with the others. Understanding the business is the minimum foundation, because every other stage starts from it.
How does work with Tog Lab start?
With a 45-minute call with someone from business and someone from engineering on our team. You come away with a candid view of what is worth building now and the next steps.
How long does it take to launch a product?
It depends on what needs to exist in the first version. In the first call we separate the essentials from what can wait and put together a staged plan, with a deliverable you approve at the end of each stage.
Do you work alongside my in-house team?
Yes. Many projects include client people in product, design or engineering. Tog Lab organizes the method and the reviews, and your team follows every decision.
Can you take over an existing system?
Yes. We start by reading the code and the documentation, map the risks and modernize step by step, with the old system running until each part is replaced.
Innovation and Service as a Software
What does Tog Lab do in innovation?
We help leadership look at the business from a different angle: research with the people who use it, design sprints to test ideas in a few days and new business model design before investing heavily.
What happens in a design sprint?
In about five days, the group defines the problem, generates solutions, picks one, builds a prototype and tests it with real people. You come away with evidence to decide the next step.
Does Service as a Software work for any business?
It works best when the client pays for work that repeats, such as analyses, reports or customer service. We assess with you where it makes sense to deliver the finished work and where software remains a tool.
Do I need a SaaS product to consider Service as a Software?
No. You can start from a service that today is done only by people and design the version in which AI agents take part in the work, with your team reviewing the result.
What is the difference between Task as a Service and Service as a Software?
With Task as a Service you buy each task completed by AI agents, such as a reconciled transaction or a resolved ticket. With Service as a Software you buy the whole service, which is made of many of those tasks. Many companies start with one task and later combine tasks into a service.
Outsourcing
Can I interview people first?
Yes. You get the profiles with experience, stack, languages and time zone, talk with each person and can run your own technical test. The engagement starts only after you sign off.
What if someone leaves or goes on vacation?
Vacations are announced in advance and covered by someone who already follows your project. If someone leaves, we present a replacement within the timeframe set in the contract. Documentation kept with the code shortens the handover.
Who reviews the code of the people placed on my team?
A Tog Lab professional oversees the engagement and reviews the code, with the experience your project calls for. You get a monthly report on what was delivered.
Do you work in English?
Yes. People placed with clients outside Brazil are fluent in English and work in English in meetings, pull requests and documentation. Spanish, Italian and German depending on the profile.
How is this different from hiring employees or contractors directly?
When you hire directly, recruiting, management, code review and replacement are on your company. With outsourcing, Tog Lab selects, oversees, reviews the code, sends the monthly report and is accountable for continuity.
AI-Accelerated
Which AI model do you use?
Whatever makes sense for the project and for your company’s rules. AI-Accelerated switches models through configuration and, in your environment, uses your key.
Does the AI have access to my data?
In your environment, AI-Accelerated runs in your account and with your key. The agents work with the data structure and with what you grant access to, and every access is logged.
Who is accountable for what goes to production?
A person. AI agents do the heavy lifting at each stage, but nothing moves forward without approval from whoever is accountable for the result.
Do I need to change the tools my team already uses?
No. AI-Accelerated works in your repository and your tools, and the rules for what the agents can do are agreed with your team.
Who owns the code and the documentation?
You do, from the first commit. Code, tests and the documents from each stage stay in your repository.
Training
Does the training use a real case from our company?
Yes. The group practices on a real challenge or workflow from your team, and the artifacts built in class stay with the company.
Can it be done remotely?
Yes. Workshops and tracks run in person or remotely, with the same material and the same exercises.
Who is the training for?
For leaders and for product, design and engineering teams. The path is built around who will take part and what the company wants to change.
Does Tog Lab follow up after the last class?
Yes. We support the first applications in the team’s workflow and review the deliverables with the participants.
Engagement and contracts
What are the ways to engage?
By project, outsourcing, AI-accelerated development in your environment and training. You can start with any of them and combine them later.
How does the contract work?
Each engagement model has its own contract, with scope, responsibilities and confidentiality. Code, documents and data belong to the client from day one.
Do you sign a nondisclosure agreement?
Yes, before any exchange of sensitive information.
Do you work with companies outside Brazil?
Yes. Tog Lab serves clients in the Americas, Europe and the Middle East, with teams in Brazil and Portugal covering business hours on both sides of the Atlantic.
Terms explained
What is AI-Accelerated?
It is Tog Lab’s development model: AI agents do the heavy lifting across seven stages, from context to infrastructure, and a person approves each stage before the next one starts.
What is a harness?
It is the structure around the AI agents: the business context, the tools they can use, the automated checks and the log of what was done. The name comes from a horse’s harness: the model provides the power, and the harness directs that power into useful work, within limits.
What are guardrails?
They are rules that limit what the agents can do, such as blocking commands that delete data or libraries without a usage license. They work as safety locks.
What is Service as a Software?
It is when the client pays for the finished work, not for access to a tool. The software, with AI agents and people reviewing, delivers the result, such as an analysis or a report.
What is Task as a Service?
It means buying the task done, one at a time. AI agents carry out the task within the business rules, a person decides the cases outside the rules, and billing is per completed task.
What is discovery?
It is the stage where you understand the problem before building: who uses it, what they need, what the business gains and what can wait.
What is a design sprint?
A method of about five days to go from a business question to a prototype tested with real people.
What is a Domain Context?
The document that gathers what you need to know about the business before building: rules, terms, risks and open decisions. Every other stage starts from it.
