JC.

Currently building @ Loopa Finance

JuaniColomboSoftwareEngineer.

I build systems that turn complex problems into reliable products.

Software Engineer focused on backend systems, cloud infrastructure and AI-powered products.
Based in Buenos Aires, Argentina.

View my workGet in touch

About

I like working where software engineering meets real-world complexity.

I’m a Software Engineer from Buenos Aires working across backend engineering, cloud infrastructure, automation and applied AI.

I enjoy designing systems, understanding how they fail, and turning ambiguous product problems into software that can actually operate reliably.

00

Years shipping

00

Live products

2021

At Loopa since

  1. 01 / Origin

    Buenos Aires

  2. 02 / Craft

    Backend & infrastructure

  3. 03 / Focus

    Applied AI

  4. 04 / Now

    Loopa Finance

What I work on

Four surfaces. One way of thinking.

01

Backend Systems

APIs, workers and services designed to stay correct when the happy path is no longer the common path.

  • TypeScript
  • Node.js
  • Python
  • REST APIs
  • Distributed systems

02

Cloud & Infrastructure

AWS systems with as few moving parts as possible — and the ones that remain are observable.

  • AWS
  • ECS
  • Lambda
  • S3
  • SQS
  • CloudFront
  • OpenSearch

03

AI / GenAI

Models placed behind real workflows: retrieval, embeddings, tools and evaluation — not demos.

  • LLMs
  • RAG
  • Embeddings
  • Vector Search
  • Agents

04

Product Engineering

Turning ambiguous product problems into software a team can operate, debug and improve.

  • System Design
  • CI/CD
  • Observability
  • Automation

Experience

Where this work lives.

  1. Aug 2021 — Present

    Current

    Senior Backend Developer

    Loopa Finance · Remote

    Building backend infrastructure, AI systems and internal products for a technology-driven litigation finance company. Work spans legal data pipelines, RAG and document intelligence, AWS, automation, and the platforms the team actually operates on. Selected for the AWS CTO Fellowship LATAM.

    • Backend architecture
    • AI / GenAI
    • AWS
    • Data pipelines
    • Automation

  2. Jan 2021 — Oct 2021

    Analytics Intern

    Accenture · Buenos Aires

    Intensive professional program focused on data analysis and machine learning. Worked with Python, pandas, NumPy and Spark, SQL / MySQL, Hadoop, and Google Cloud Platform for managing and scaling projects.

    • Python
    • Data analysis
    • Spark
    • SQL
    • GCP

  3. Jan 2020 — Jan 2021

    Front End Developer

    Pipet · Buenos Aires

    Contractor on a web and mobile product. Built interfaces and services with React and Node.js, using Firebase for realtime data and HTML, CSS and JavaScript for the client.

    • React
    • Node.js
    • Firebase
    • JavaScript

Selected work

Selected systems and products

Project 01

AI Knowledge Infrastructure

A document intelligence platform that transforms unstructured legal information into searchable, AI-ready knowledge.

  • RAG
  • AWS
  • Vector Search
  • Distributed Processing

Architecture

Documents
Object Storage
Queue
Workers
Embeddings
Vector Search
LLM Applications

Project 02

Legal Data Infrastructure

Systems for collecting, normalizing and processing large-scale legal data across multiple jurisdictions.

  • Python
  • Automation
  • Data pipelines
  • Backend systems

Architecture

Sources
Crawlers
Processing
Storage
APIs
Internal products

Project 03

AI-Powered Product Systems

Applied AI systems that assist users with document understanding, extraction, classification and decision-support workflows.

  • LLMs
  • Agents
  • RAG
  • Product Engineering

Architecture

User
Application
AI orchestration
Retrieval
Tools
Structured Data
LLM
Structured Result

Live product 04

GuitaApp

A personal finance product for accounts, credit cards, installments, recurring payments and monthly projections — with an AI assistant on top.

guitaapp.com
  • Product
  • React
  • AI assistant
  • Personal finance

Architecture

Accounts
Cards & cuotas
Ledger
Projections
AI assistant
Dashboard

Live product 05

F1 Prode

A Formula 1 prediction platform where people pick race results, create private championships and compete for points across GP, sprint and qualifying.

f1prode.com
  • Product
  • Next.js
  • Live data
  • Rankings

Architecture

Race weekend
Predictions
Scoring
Leagues
Rankings

Engineering principles

A few things I keep coming back to.

01

Simple beats clever.

The system you can explain at 2am is the one that survives.

02

Reliability is a feature.

Users do not experience your architecture. They experience whether it works.

03

Architecture starts with constraints.

Latency, cost, failure domains and the shape of the data come before the diagram.

04

AI is useful when it solves an actual problem.

A model is a component. The product is the workflow around it.

Writing

Notes on engineering

Future technical writing. These are titles I intend to publish — not articles that already exist.

  • Designing reliable asynchronous workers

    Queues, visibility timeouts, and the difference between retryable and poison work.

    Coming soon

  • What I learned building RAG systems

    Chunking, retrieval quality, and why the index is more important than the prompt.

    Coming soon

  • Queues, retries and idempotency

    How to make work safe to run twice — and expensive to run wrong.

    Coming soon

  • Building AI features that survive production

    Evaluation, fallbacks, and keeping a model from becoming a single point of failure.

    Coming soon

When I’m not building software

Travelling·Snowboarding·Running·Learning how systems work·Travelling·Snowboarding·Running·Learning how systems work·

Contact

Let’s build something interesting.

I’m always interested in ambitious engineering problems, AI products and conversations about building great software.

Email meLinkedInGitHub