Attitude Informática Ltda · Rio de Janeiro, Brazil

Software engineering that becomes a competitive advantage.

We design the architecture, build the team and ship the system. For over 20 years, across telecom, healthcare, education, retail, social security and industry associations — with the discipline of someone who has worked inside CMMI level 5 operations and offshore delivery — today on AWS, with AI inside the delivery process.

Talk about your project See what we do
20+ years Building, architecting and leading software delivery
AWS Certified Solutions Architect, with architecture running in production on AWS
CMMI level 5 Hands-on experience inside CMMI and ISO certified development processes
Offshore Leading distributed teams with direct client-facing responsibility
ITIL · COBIT Certifications and Scrum practice backing the process, not just the pitch

What we do

Three ways to work with us.

Not every problem starts in the same place. Sometimes the technical decision is missing; sometimes the hands are; sometimes the whole product is. We work on all three — and we will tell you plainly when yours is not one of them.

01

Architecture and technical leadership

For when technology has stopped keeping up with the business: legacy blocks every release, nobody can say what a change costs, and the team grew without a technical backbone.

  • Architecture assessment and design (DDD, Clean Architecture, microservices)
  • Legacy modernisation and migration to fully containerised environments
  • Extracting services from monolithic codebases without stopping operations
  • Building, organising and mentoring engineering teams
  • Standards, metrics and IT governance (CMMI, ITIL, COBIT)
02

Squads and software delivery teams

For when the plan is clear and what is missing is capacity to execute. We assemble the team, take on the backlog and deliver in short cycles, with scope, cost and schedule visibly under control.

  • On-demand teams: full-stack, QA and DevOps
  • Incremental delivery in two to four week sprints
  • Requirements elicitation and scope, cost and schedule control
  • Automated unit, integration and end-to-end testing
  • Continuous integration and delivery, with a staging environment
03

Custom platforms

For when off-the-shelf software does not solve it. We build the product from scratch — from interviewing the people who will use it through to production, with the cloud account in the owner's name.

  • Multi-language, multi-role and multi-tenant web applications
  • Payments via Pix, bank slip and card, with automated reconciliation
  • Documents generated at scale: certificates, badges and reports
  • AWS infrastructure designed around data protection law from day one
  • Ongoing support, evolution and knowledge transfer

Cloud and AWS

AWS architecture sized for the problem — not for the service catalogue.

A badly designed cloud is expensive twice over: in the invoice that arrives, and in the effort of whoever has to change it later. We design for the real volume, with the growth path written down before it is needed.

  • Legacy migration to a fully containerised environment
  • Containers on ECS, with versioned images in ECR and automated deployment
  • Cost engineering: size for actual usage and switch off what is not running
  • Data protection considered up front — region, retention and access decided before the code
  • Observability, backup and restore tested rather than assumed
A row of server racks in a data centre

A real example: the ABRACO platform in production

Diagram of the ABRACO platform production architecture on AWS. The text just below describes the same flow.

Drag the diagram sideways to see the rest.

How to read it: the flow runs left to right and no line crosses another — that is layout, not luck. In green, the path of a request: the user resolves the domain at Route 53, which sends media to CloudFront and the application to the load balancer, and the balancer serves the Next.js application running on ECS Fargate. Dashed, what the application consumes: S3, RDS MySQL, SES, Secrets Manager and CloudWatch, plus the Banco do Brasil (bank slip) and PayPal (card) APIs, which sit outside the cloud. Above and below ECS, the two pieces that serve no request: ECR, holding the released image, and the EventBridge rule with Lambda that shuts the service down out of hours.

About the highlighted node: the client's environment does not need to exist at three in the morning. An EventBridge rule calls a Lambda function that scales the ECS task count to zero at night and brings it back in the morning — so the cloud bill covers the hours when someone is actually using the system. That kind of decision never shows up in a proposal diagram, and always shows up in the invoice.

What is deliberately not here yet: WAF and an ElastiCache layer are designed and prioritised, but they go in when the volume justifies them. A conference with 500 to 1,500 attendees does not pay for a cache cluster — and saying so to the client is part of the job.

About the icons: these are the official AWS Architecture Icons, used here unmodified and in an architecture diagram — the purpose Amazon publishes them for. AWS, Amazon Web Services and the service names are trademarks of Amazon.com, Inc. or its affiliates. The reference is descriptive and does not imply partnership, sponsorship or endorsement by AWS.

How we work

Genuinely agile: short cycles, negotiated scope, no surprises at the end.

Changing requirements are not an accident — they are normal on any project that runs longer than a month. Our process assumes that from the contract onwards: every change goes in writing, with its schedule and cost impact on the table, and the decision to proceed is yours.

  1. Planning

    A joint session to settle objectives, priorities and scope. It produces the initial backlog and the acceptance test plan.

  2. Sprints

    Two to four week iterations. Each one ends with working functionality, not with a progress report.

  3. Reviews

    At the end of every sprint, a demonstration of what was built and immediate feedback — with you watching the system run.

  4. Retrospectives

    The team assesses what worked and what got in the way, and adjusts its own process before the next sprint.

Transparency Progress is visible every sprint, without waiting for a monthly report.
Flexibility Changes are accepted and priced, rather than refused or billed quietly.
Value sooner The functionality that matters goes into use before the project ends.
Quality Frequent testing and constant feedback, instead of a last-minute acceptance rush.

Artificial intelligence

AI where it pays for itself.

We do not sell AI as a category. We sell three concrete uses — two of them inside the client's product, and one inside the way we deliver. That last one is in daily use, not in pilot.

Close-up detail of a printed circuit board
01

AI-assisted engineering

The way we deliver. Agents specialised in architecture, code, review and testing work against a rule set versioned alongside the project — so the same standard applies to everyone, and nothing is lost when the person at the keyboard changes.

  • Project standards as versioned text, not as a verbal agreement
  • Automated review before human review
  • Documentation produced alongside the change
  • A shorter cycle without giving up the human reviewer at the end
02

GenAI inside your product

For when AI is a feature rather than an internal tool: search that understands the question, an assistant that answers from your own data, extraction from documents. Built on Amazon Bedrock, inside your account.

  • RAG over the client's own data, always citing the source of the answer
  • Amazon Bedrock in your AWS account — your data never leaves your perimeter
  • Guardrails, cost ceilings and a log of everything that was asked
  • Evaluated against real cases before launch, not demo-day impressions
Amazon Bedrock
03

Process automation

The repetitive work that currently consumes a whole person: reading a document, classifying it, checking it against the system, routing it. AI applied to back-office process, with a human deciding the exceptions.

  • Data extraction from PDFs, invoices, bank slips and forms
  • Classification and triage of requests arriving by e-mail
  • Automated checking against business rules
  • An exception queue for human review — the AI does not decide alone

Platforms

An event and conference platform, built from scratch.

Built for ABRACO, the Brazilian Corrosion Association, after off-the-shelf tools failed to handle the organisation's process.

The system covers the whole event, from the first registration to the certificate: sign-up and payment, two-stage technical paper submission and review, sponsors, exhibitors and the operation on the day — badges, barcode check-in and the income-versus-expense statement.

It has been multi-language since day one, because the association runs international conferences, and it runs on AWS with data protection considered in the design rather than patched in later.

Next.js TypeScript React Prisma MySQL AWS S3 · SES Docker i18next Jest · Playwright

What the platform handles

  • Registration and roles. Administrator, attendee and finance, with pricing that varies by membership category.
  • Finance. Bank slip, card and Pix, with reconciliation file processing and payment settlement.
  • Technical papers. Abstract submission, two-stage committee review and automated author notifications.
  • Certificates. Per-event layouts, automatic issuing after attendance and payment, reissue and download.
  • On-site operation. Badge printing and attendance lists via barcode scanning.
  • Sponsors and exhibitors. Booth selection, automatic fee calculation and management of member companies.
  • Reporting. Attendees, registrations, payments, attendance and the event's financial statement.

Where we have been

Sectors that need a working system, not a good-looking prototype.

Dense business rules, integration with an old system nobody is allowed to switch off, audits and statutory deadlines. That is the kind of environment that shaped the way we work.

Telecommunications Fraud detection, device management and blocking, tax and fiscal systems
Healthcare Development and maintenance of critical modules on an offshore project
Education IT leadership, academic systems and campus infrastructure
Retail Corporate systems and end-to-end process integration
Government and social security Projects with statutory obligations and regulatory deadlines — delivered on time
Entertainment Systems supporting production and operations
Associations and institutions Platforms for events, conferences, membership and certification
Consulting and enterprise IT Level 3 support, technical leadership and architecture on large accounts

Projects and organisations along the way

  • HPE Pointnext
  • IBM Brazil
  • EDS Brazil
  • Oi
  • Vivo
  • Embratel
  • Laureate · IBMR
  • Dataprev
  • IBGE
  • Lojas Marisa
  • Contax
  • Conspiração Filmes
  • Estácio de Sá
  • Telelistas
  • Lojas Brasileiras
  • ABRACO

Organisations the professionals at Attitech have worked for over more than 20 years, either on projects contracted directly by Attitude Informática or in previous roles as consultants and technical leads. The names above belong to their respective owners; mentioning them here is descriptive and does not imply partnership or endorsement.

About us

Small by choice, senior by necessity.

Attitude Informática exists to solve the kind of problem that does not fit in a support ticket: choosing the right architecture, building a team that can stand on its own, and getting a difficult system into production. The person running the project is the person you talk to.

A full-stack developer and technical lead by training, now focused on software architecture and on building teams. The core strength is the ability to deliver — and to build an engineering organisation so closely aligned with business strategy that it becomes a competitive advantage in itself.

Working across very different sectors, combined with hands-on exposure to international quality models (CMMI) and offshore projects, is what makes it possible to walk into a new context and quickly recognise what has already gone wrong before.

Education BSc in Computer Science (Universidade Gama Filho) and an MBA in Software Engineering (UFRJ)
Certifications AWS Certified Solutions Architect – Associate; ITIL v3 Foundation, COBIT 4.1 and Scrum Fundamentals Certified; volunteer project manager at PMI Rio
Languages Native Portuguese and advanced English, with experience working directly with clients and teams abroad

Contact

Tell us the problem. The first conversation is free.

A technical assessment, a squad to assemble or a product from scratch — start by describing the situation and we will come back with an honest read on what is feasible, how long it takes and what it costs.

WhatsApp+55 (21) 99782-0272 E-mailcontato@attitech.com.br LinkedInMichel Gomes