Guilherme Soares Sousa Silva
Guilherme Soares Sousa Silva

Senior Software Engineer

Actively looking · Member since 8 Oct 2026
Location
São Luís, Brazil
Desired salary
4000–7600 USD/monthly
Work preference
Remote Only / Full Time, Contract
Experience level
Senior

About

Professional summary

Senior Software Engineer with 6+ years of experience in SaaS, financial services, public digital services, e commerce and legal systems. I work in product engineering with a backend focus, building APIs, integrations and production services with Go, Node.js and TypeScript. I also build automations, orchestration, data processing and pipelines with Python, and apply AI to real products using LLMs, RAG, embeddings, vector search and agents. In a real estate SaaS with 96k+ registered users, I cut lead processing time by about 45% and shipped LLM features integrated with product data and business rules. Currently at Hitss Brasil on a mission critical digital services platform for a state traffic authority, using Go, Node.js, Kafka and PostgreSQL. Bachelor's degree in Computer Science. Previous experience with PHP/Laravel and Java.

Notice period: 2 weeks

Skills

10 capabilities

Tech stack & tools

Working toolkit

Experience

Career history

Senior Software Engineer Hitss Brasil

Rio de Janeiro, RJ (Remote). Challenge: Production digital services platform for a state traffic authority, supporting citizen services and mission critical internal operations, with complex business rules, multiple integrations and high requirements for security, traceability and reliability. Actions: Designed and implemented product backend services and REST APIs in Go and Node.js for critical production flows, running as Docker containerized services. Restructured synchronous operations into asynchronous processing with Kafka, isolating slower dependencies and increasing reliability. Investigated recurring failures between services and improved validation, retry and error handling. Optimized PostgreSQL data access flows, reducing unnecessary queries and improving response time for higher volume operations. Delivered features end to end, from business rules analysis and technical design to automated tests, integration and production validation. Results: About 35% lower average execution time in the optimized flows, about 30% fewer failures during requests and about 40% fewer defects reaching acceptance testing.

Senior Software Engineer Apresenta.me

Brazil (Remote). Contract engagement, partly overlapping with full-time role. Challenge: Product SaaS for the real estate market with more than 96,000 registered users, covering CRM, property listings, leads, integrations and automations that support the daily operations of real estate agencies. Actions: Built and evolved REST APIs and product backend services in Node.js and TypeScript. Restructured lead processing routines with asynchronous execution and background services, removing heavy operations from synchronous requests. Automated lead and CRM operations that previously required repetitive user actions, and applied AI, LLM and RAG features integrated with product data and business rules to automate and qualify lead flows. Investigated integration bottlenecks, improved validations and failure handling, and delivered changes incrementally in production with automated tests and controlled releases. Results: About 45% lower lead processing time in the affected flows, about 50% fewer manual steps in selected flows and about 30% fewer recurring processing errors.

Senior Software Engineer VRS Components

Brazil (Remote). Challenge: Digital transformation of the commercial operation of an electronic components distributor, previously fragmented across spreadsheets, messages and manual catalog and order controls, consolidated into a single product system. Actions: Built the product backend in Go and Node.js, connecting catalog, orders and administrative operations through structured REST APIs. Replaced manual order control steps with automated flows, and designed validation and business rule services that blocked incomplete or inconsistent orders. Built Python automations and data processing routines supporting catalog and order operations. Consolidated product, order and administration data into one application as the central operational source, and handled requirements, domain modeling, architecture decisions, integrations and production delivery. Results: About 60% less manual intervention in the digitized processes, about 35% less operational rework, and the main order processing flow reduced from about 20 to 8 minutes.

Software Engineer CD2 Retail Tech

Brazil (Remote). Challenge: Financial product integrated into a retail operation with more than 5,000 stores, a critical system where data inconsistencies, integration failures or incorrect authorization directly affect financial and operational transactions. Actions: Built Node.js backend services and REST APIs for financial transactions and integrations used by a network of more than 5,000 stores. Implemented business validations that blocked invalid or inconsistent data from reaching financial processing. Moved non blocking operations to background processing, and implemented authentication and authorization rules for sensitive financial operations, strengthening access isolation between user profiles and protected resources. Expanded automated test coverage of critical flows and improved failure handling in integrations, reducing cases that required manual operational correction. Results: About 30% fewer recurring data inconsistencies, about 40% lower response time in selected API flows, and test coverage of critical flows raised from about 45% to 75%, reducing regressions in later deliveries.

Software Engineer JV Advogada

São Paulo, SP (Remote). Challenge: Internal systems for a legal operation that relied on spreadsheets, documents, emails and manual controls to run administrative and procedural workflows. Actions: Built internal applications and backend services in Go, replacing flows spread across spreadsheets, emails and manual document controls. Designed REST APIs and business rules to centralize administrative information and process execution. Turned recurring manual procedures into structured digital workflows, and built Python automations and document processing routines for recurring administrative and case related tasks. Implemented role based access control and audit trails for sensitive internal information. Worked end to end, from requirements and process mapping to architecture, development, deployment, user validation and continuous evolution. Results: About 50% less duplicated data entry, about 40% lower average execution time of automated routines, and higher traceability of actions and document access, with operational information consolidated in one central system.

Software Engineer Betabit Dev

Curitiba, PR (Remote). Challenge: Software house delivering projects in e commerce, real estate, CRM, compliance and administrative systems, requiring constant adaptation to different domains, legacy systems and business rules. Actions: Built backend services, REST APIs and business rules, mainly in Node.js, for e commerce, CRM, real estate and administrative products. Integrated external systems and internal modules, replacing duplicated data entry and reducing manual handoffs between applications. Restructured critical backend logic in a compliance platform with complex validation rules, maintained and evolved legacy PHP and Laravel systems by gradually isolating new features into better defined components instead of disruptive rewrites, and expanded automated tests in critical modules. Supported code reviews, solution design, task breakdown, technical decisions and delivery coordination in selected projects. Results: About 30% fewer recurring validation defects in the compliance platform, safer releases, less rework caused by regressions and higher release confidence in critical modules.

Software Engineer Confidential Client

Brazil (Remote). Challenge: Independent software development for small businesses, covering websites, e commerce, integrations and custom web solutions for specific commercial needs. Actions: Built web applications and backend features with Node.js and PHP, turning business requirements into features used in production. Developed and customized Magento online stores, including checkout, catalog, integrations and operations adjustments for different business scenarios. Automated repetitive catalog and administration routines, built frontend interactions with JavaScript and jQuery integrated with backend services and business rules, and diagnosed and fixed performance and functional issues in existing sites. Managed delivery directly with clients, from requirements clarification and implementation to testing, deployment, production fixes and post launch adjustments. Results: About 30% less manual content management work and about 40% faster load times in the optimized projects, with performance and functional fixes and post launch adjustments delivered directly to clients.

Education

Learning history

Faculdade Pitágoras de São Luís

Bachelor's Degree, Computer Science

This professional hasn’t added portfolio projects yet.

This professional hasn’t listed any services yet.

Jobs Talent AI Tools Salaries
Menu