Understand
Clarify what needs to happen, who uses it, what decisions matter and what success looks like.
Full-stack development, automation, integrations and pragmatic architecture for product and operations teams.
0+
years of experience
0
layers covered
0
working languages
0
owner end-to-end
I can take a public-facing site, a product feature, an internal workflow or an AI-powered idea and turn it into a working, maintainable slice.
// ai-powered
LLM integrations, agents, MCP-style tool use and reviewable AI steps embedded inside practical products.
Landing pages, personal sites and marketing surfaces that look intentional, load fast and explain the offer clearly.
Features that need real state, forms, APIs, databases, roles, validation and deployment context.
Operational screens for teams that need to see, filter, approve and act on their data without fighting spreadsheets.
Workflows that connect tools, remove repeated handoffs and make the next action traceable.
Pragmatic structure for codebases that need clearer boundaries, shared packages, better contracts or safer iteration.
Real professional work first, then intentionally labeled portfolio builds that show the direction of my freelance offer.
Large React systems
Frontend pieces of a large enterprise platform with independently deployed applications and shared libraries.
context
Enterprise teams need UI that can evolve across teams without each feature becoming a local exception.
architecture
Independently deployed React microfrontends with shared packages, typed API contracts and review-heavy delivery.
build
React and TypeScript screens, Module Federation integration, custom hooks and shared UI behavior inside a large platform.
Shows I can work inside serious codebases, collaborate through review and ship maintainable features.
Workflow automation
A practical automation project for keeping technical documentation closer to the work that changes it.
context
Documentation falls behind when updates depend on repeated manual coordination.
architecture
Event-triggered automation with APIs and generated updates connected through a traceable workflow.
build
n8n workflows connected triggers, APIs and generated updates into a reviewable documentation flow.
Shows how I identify repeated work, connect tools and make the result inspectable.
AI-assisted operations
A demo for teams handling invoices, emails, PDFs or forms that need structured, reviewable data.
context
Teams copy information from documents into systems by hand, then lose time checking what changed.
architecture
AI extraction is bounded by schema validation, human review, audit trail and export-ready records.
build
A planned Next.js/NestJS demo for invoices, emails, PDFs or forms that need structured, reviewable data.
Shows AI as one layer of a product: helpful for extraction, bounded by validation and human control.
Looks simple, but...
A personal site used as a live case study in positioning, bilingual content, motion, components and performance.
context
A portfolio looks like a small page, but it has to explain a broad offer fast, feel credible and work for both technical and non-technical clients.
architecture
Static-exported Next.js with typed content, bilingual routes, componentized sections and Cloudflare Pages deployment.
build
Curated 21st.dev-inspired interactions, custom cards, motion fallbacks and a visual system built as production code.
Shows I can build frontends that feel intentional, not just functional, while keeping the engineering maintainable.
These are the tools and practices I can bring together depending on what the client actually needs.
I care about useful systems: interfaces that explain themselves, backend boundaries that are hard to misuse, and automation that reduces work without hiding control.
Bringmeagoalandconstraints.I'llfigureouttherest.
Frontend,backend,automations,AIworkflows,integrations—Iworkacrossallofit.Whensomethingnewcomesup,Istudyitanddecideifitactuallyfits.
Idon'tleavethingshalf-done.Ibuildwhat'sneeded,andstayuntilitworks.
The process adapts to the project, but the discipline stays the same: clarify the goal, map the moving parts, build the useful core, then improve from feedback.
implementation path
Each step reduces ambiguity until there is a working slice with clear ownership and reviewable behavior.
Clarify what needs to happen, who uses it, what decisions matter and what success looks like.
Identify screens, data, integrations, permissions, edge cases and manual steps that can become software.
Choose the first version that creates value without overbuilding the whole future system.
Implement the frontend, backend, validation, integrations and deployment details needed for the slice to work.
Test the behavior, keep important actions reviewable and iterate from real usage.
I combine frontend craft, backend implementation, integration work and AI/automation direction.
I am Santiago Varacca, based in Buenos Aires and working with remote teams. My strongest production base is React, TypeScript, microfrontends and enterprise UI work, with backend experience across Node.js, NestJS, REST, gRPC and databases.
The direction now is applied AI: not as a label, but as part of useful software. If a workflow benefits from extraction, routing, agents, automation or model-assisted steps, I can design that layer around reviewable product behavior.
Frontend that looks and feels intentional
Backend and integration work that supports the product
Architecture decisions that keep the system maintainable
AI and automation used where they create real leverage
Tell me what you are trying to build. If it needs software, systems or AI, we can map the right solution.