nicolas.souza

STRATEGY · TECHNOLOGY · INNOVATION

UNDERSTAND.
BUILD.
CREATE VALUE.

3 million and 500 thousand Brazilian reaisIN RETURNS GENERATED THROUGH AI INITIATIVES

I'm Nico.
Head of Technology and Innovation.
Technology Executive.

I connect business strategy, people and technical execution to turn technology into results. The agentic AI initiatives I lead have freed up more than 200 hours of work and generated BRL 3.5 million in returns.

Explore my work
NICOLAS SOUZA / TECHNOLOGY
Illustrated portrait of Nicolas Souza in a black shirt with golden lighting.
PEOPLE + CONTEXT + TECHNOLOGY
01Problems understood02Informed decisions03Creative solutions04Products in use05Measurable impact

01 / WHAT I BUILD

REAL PROBLEMS.
SOLUTIONS IN USE.

I work with engineering teams to turn the needs of leadership, sales, operations and finance into working solutions.

01
PRODUCT MINDSET

Solve problems. Win hearts and minds.

I take ownership of the problem, question the request and build creative solutions that people understand, value and want to use.

Product visionApplied creativityValue for people
Expand context

Look beyond the request

The request starts the conversation. I investigate what lies behind it: user needs, business goals and constraints. I take responsibility for identifying the right problem.

Bring a solution

I challenge assumptions, connect perspectives and propose creative approaches. I turn an idea into something tangible that people can try, discuss and improve.

Win hearts and minds

A solution needs to make sense and make people want to use it. I aim for clarity that builds trust, thoughtful experiences and lasting value that supports adoption.

02
APPLIED ARTIFICIAL INTELLIGENCE

AI inside the workflow.

Agents, procedures and integrations that expand team capacity.

Claude CodeCodexMCPs & skills
Expand context

The problem

Teams spend part of their effort gathering information, rebuilding context and repeating analysis.

My role

I apply AI to research, development and delivery. Skills structure procedures; MCPs connect tools and sources of context.

How I evaluate it

I establish a baseline and track time, cost and quality. Technology must generate returns, increase productivity or reduce waste.

03
DATA & LEADERSHIP

Clarity to decide. A team to deliver.

Engineers and data analysts working with business context.

BITechnical leadershipOperations
Expand context

The problem

A metric only helps when the team understands the question, the business rule and the decision it informs.

My role

I lead engineering and BI. I contribute to our Lakehouse and Warehouse development, connecting reliable data to decisions about performance, profitability and efficiency.

What guides the work

Traceable sources, explicit criteria and a shared understanding of what the data supports.

Areas of work. Open each one to explore the context and approach.

02 / HOW I WORK

FIRST, I UNDERSTAND.
THEN, AI ACCELERATES.

Before writing code, I investigate the problem, its current cost and the expected outcome. Then I coordinate people and agents to build, launch and measure.

FROM PROBLEM TO EVIDENCETHE COMPLETE PROCESS
Human diagnosis, planning, AI-assisted development and validation Rework, scattered data and slow decisions go through Nicolas’s analysis, value definition and prototyping before agent execution and outcome comparison. Rework Scattered data Slow decisions My analysisROOT CAUSE VALUE MAPPain + costPriority + goal SOLUTION DESIGNPROTOTYPE + RULES Claude CodeCodexMCPs · n8nTEAM + AGENTS ROIVALIDATIONBEFORE × AFTER HUMAN DIAGNOSIS & DIRECTIONEXECUTION & VALIDATION

01 / Study the workflow and identify the root cause.

02 / Record the baseline and define the expected value.

03 / Make the solution tangible and document decisions.

04 / Coordinate the team and agents around verifiable deliverables.

05 / Compare gains, investment and adoption against the baseline.

Illustration of the method. Return measurement depends on the data available for each project.

HOW I WORK

Problem solving, research and execution.

From the problem to an application in production.

01 / PROBLEM SOLVING

Investigate the cause.
Define the problem.

I analyze the process with the people doing the work, compare their accounts with data, and identify rework, failures and costs. I define the problem before deciding what to build.

Current process, root cause and measurable impact.

TOOLSSlackNotion

DELIVERABLE / DIAGNOSIS

Problem definition
CONTEXT
Who is affected, during which task, and how often.
EVIDENCE
Operational data, examples of failures and limitations of existing systems.
BASELINE
Time, cost or error rate that the solution needs to improve.

02 / RESEARCH & PRODUCT DISCOVERY

Research the context.
Test the hypotheses.

I interview users, observe their work and investigate existing solutions. I compare alternatives against business and technical constraints to decide what needs validation first.

Interviews, data analysis and hypothesis testing.

TOOLSNotionClaude Code

DELIVERABLE / DISCOVERY

Evidence for decisions
RESEARCH
User needs, existing tools and gaps in the process.
HYPOTHESES
Which change could solve the problem, and how to test that relationship.
PRIORITIZATION
Expected impact, effort, dependencies and risks of the initial scope.

03 / SOLUTION DESIGN

Design the product.
Specify the implementation.

I turn hypotheses into workflows, prototypes and requirements. With the team, I define the architecture, integration contracts and business rules that guide development.

Prototype, technical specification and acceptance criteria.

TOOLSFigmaNotion

DELIVERABLE / SOLUTION SPECIFICATION

From discovery to a technical plan
PRODUCT
User journey, prototype and validation of core tasks.
ENGINEERING
Data model, APIs, permissions and failure handling.
PLAN
Scope divided into testable deliverables, with dependencies and acceptance criteria.

04 / TECHNICAL EXECUTION WITH AI

Build with AI.
Review with engineering rigor.

I work with engineers, data analysts, Claude Code and Codex on implementation. I use skills and MCPs to give agents context, integrate tools and accelerate code, tests and documentation.

Implementation, code review and automated testing.

TOOLSCodexClaude Coden8n

DELIVERABLE / IMPLEMENTED SOFTWARE

Executing the technical plan
DEVELOPMENT
Interfaces, services, queries and integrations within the defined scope.
APPLIED AI
Agents with bounded tasks, project context and verification criteria.
QUALITY
Code review, behavior tests and permission checks before deployment.

05 / VALIDATION IN PRODUCTION

Put it into operation.
Measure the outcome.

I validate the application with users, monitor errors and compare metrics against the baseline. I use the results to improve the product and plan the next development cycle.

Real usage, reliability and return on investment.

TOOLSSlackNotion

DELIVERABLE / VERIFIABLE RESULTS

What to track after deployment
OPERATIONS
Task completion, failures, performance and user difficulties.
IMPACT
Changes in time, cost and errors compared with the previous process.
ITERATION
Fixes and new priorities based on actual usage and observed results.

03 / WORKING NOTES

JUDGMENT BEFORE
THE TOOL.

Principles that guide my work with products, data and artificial intelligence.

WORKING NOTE / 01

Context is part
of engineering.

A good requirement connects the request to the real need. That clarity makes it possible to build and then verify whether it worked.

PRODUCT & PEOPLE
WORKING NOTE / 02

Automation needs
clear boundaries.

Access to a tool is just the beginning. The workflow must make explicit what an agent can do and when human review is required.

AI & OPERATIONS

04 / CONNECTIONS & KNOWLEDGE

KNOWLEDGE GROWS
WHEN WE SHARE IT.

A good conversation can change the question. A new perspective can change the path. I take part in these spaces to listen, exchange experiences and bring new possibilities into my work. Learning together helps us go further.

7 EVENTS · 11 MOMENTS

G4 Academia de IA — illustrated event photo 1

01AI & business

G4 Academia de IA

At G4’s AI academy, I deepened my use of GPT models, Claude and Manus, alongside n8n and Maker automations for process improvement. I exchanged ideas with business owners from different industries and suggested initiatives to improve their internal operations.

TOTVS Universe — illustrated event photo 1

02Technology & ecosystems

TOTVS Universe

At TOTVS Universe, I met with TOTVS’s logistics team and leadership. I learned about products related to AWS and end-to-end logistics integration, explored solutions and expanded my network in the sector.

Conferência de Tecnologia de Florianópolis — illustrated event photo 1

03Community & knowledge

Conferência de Tecnologia de Florianópolis

At TDC in Florianópolis, I took part in forums and exchanges with technology leaders. The experience combined learning from other professionals with networking around the challenges of managing technology in business.

TDC — Executivos e Diretores de Tecnologia — illustrated event photo 1

04Leadership & strategy

TDC — Executivos e Diretores de Tecnologia

I joined forums with directors and executives from companies including Stone and Sicoob. I learned from their experiences, exchanged perspectives on technology leadership and built connections with professionals facing these challenges firsthand.

ComexTech — illustrated event photo 1ComexTech — illustrated event photo 2

05Connections & innovation

ComexTech

I represented Suntrans as Head of Technology at ComexTech. I explored industry solutions, met logistics professionals and discussed operational challenges, building connections between technology and international trade.

G4 Traction — illustrated event photo 1G4 Traction — illustrated event photo 2G4 Traction — illustrated event photo 3

06Management & growth

G4 Traction

G4 Traction gave me a 360-degree view of business management, from financial flows to people management and growth. The program broadened my understanding of the business end to end and how technology decisions connect with its different functions.

AI Festival SP — illustrated event photo 1AI Festival SP — illustrated event photo 2

07Future & possibilities

AI Festival SP

At AI Festival SP, I attended creative AI talks involving Volkswagen and Claude and connected directly with Henrique Saveli. It was an opportunity to discover other approaches to applying AI and speak with people working in the ecosystem.

Illustrations based on my event photos. Explore the gallery and open each image to see it in detail.
Illustrated double portrait of Nicolas Souza: working and facing the camera, with golden lighting.
FROM OPERATIONS TO STRATEGY. STILL HANDS-ON.

05 / WHAT DRIVES ME

THE NEXT CHAPTER
HAS YET TO TAKE SHAPE.

That is what I work on.

I see a future where companies can accomplish what today feels too ambitious for their teams. Where artificial intelligence expands the human capacity to decide, create and execute. Where important ideas no longer have to wait for spare time.

I lead technology and innovation initiatives to give shape to that possibility — through applied artificial intelligence, software, automation and business intelligence.

My work begins where the question shifts from “how do we do this today?” to “what has now become possible?”.