Daniel Scheidemantel Camargo

Capellaris AI (SaaS, self-funded)

March 2025 — Present

• Designed, built and operate a production-grade, LLM-agnostic multi-tenant SaaS platform end-to-end — covering marketing, finance and legal automation use cases for enterprise customers.
• Full-stack delivery: Python/FastAPI backend, Node.js middleware layer, React front-end, PostgreSQL + pgvector, Redis, Supabase, Cloudflare WAF; GCP infrastructure (CloudRun, Firebase); OAuth 2.0, IAM, secrets management and tenant isolation.
• Built a reusable agent-orchestration layer (CrewAI + MCP) with embeddings, hybrid search (pgvector + BM25), chunking, reranking and tool-calling; integrated Anthropic Claude, OpenAI and Gemini APIs behind a unified multi-LLM abstraction.
• Implemented full MLOps lifecycle: prompt versioning, model evaluation, observability (Cloud Logging) and continuous delivery (GitLab CI/CD).
• Pilot enterprise customer live in production; direct technical account management — requirement scoping, integration support and product iteration based on customer feedback.

Mercado Livre (e-commerce, logistics)

August 2025 — Present

• Design, build and operate internal platform APIs for large-scale logistics operations serving millions of daily transactions; cross-functional Scrum team of 7, on-call rotation, incident management and post-mortems.
• Develop Python (FastAPI) and Go services; contribute to React-based internal tooling; own full production lifecycle on containerised multi-cloud infrastructure (Docker, AWS/GCP via MELI’s internal Fury platform).
• Integrate Claude and OpenAI APIs into production workflows as autonomous decision-support agents for logistics capacity planning; mentor team members on AI integration patterns and backend best practices.
• Build and maintain CI/CD pipelines (GitHub Actions) with automated testing, observability (structured logging, metrics, alerting) and progressive rollout strategies.
• Design ETL pipelines (Python/Pandas/Polars on BigQuery) feeding demand forecasting and capacity planning simulators used by commercial and operations teams for high-volume, data-driven decision-support.

Buscaônibus / DeÔnibus / Travelier (travel, meta-search)

April 2021 — August 2025

• Primary owner of a high-traffic travel search and ticketing platform processing millions of passenger queries: design, development, maintenance and on-call support.
• Operated and monetised passenger demand and pricing data pipelines — aggregating real-time availability, route coverage and fare data across hundreds of bus operators to power commercial analytics and revenue optimisation.
• Led full platform modernisation: Python 2.7/App Engine to Python 3.12/FastAPI/CloudRun (Docker), cutting average API latency by 25% and GCP infrastructure cost by 40%; directly improved search response quality and conversion.
• Designed and shipped a route intelligence and demand analytics layer (BigQuery + Looker Studio) used by business stakeholders to evaluate route performance, passenger volume trends and pricing opportunities.
• Built Gemini-based AI agents for internal operations and demand-signal processing via reusable backend APIs.
• Mentored junior engineers on Python best practices, API design and cloud-native deployment.

Portobello Group (retail, ERP, CRM, data)

December 2019 — April 2021

• Built Python/Django backend integrations between Salesforce, on-prem ERPs and downstream microservices; designed data pipelines for the commercial analytics stack.
• Operated Kubernetes CI/CD pipelines (Rancher + Jenkins) with Docker; maintained MSSQL Server (job scheduling, data pipeline architecture).
• Led technical design and stakeholder support for company-wide Salesforce rollout; developed a full financial simulator embedded in Salesforce with a Python backend.

Stormtech (consumer loans, regulated)

August 2018 — November 2019

• Developed Python automations and data pipeline integrations with a PHP monolith in a regulated financial environment.
• Built and operated an AWS-native platform (EC2, Lambda, Step Functions, DynamoDB, RDS-MySQL, Route53); tracked and resolved production incidents in a fast-paced support context.

ReciclaComigo (ESG, waste management startup)

July 2016 — February 2018

• Designed, prototyped and developed a Reverse Vending Machine using MQTT, AWS IoT Core, Raspberry Pi, Arduino, Computer Vision (OpenCV) and Python.
• Connected to an AWS architecture (EC2, Lambda, RDS-MySQL) to manage machine operations and collect data from users and collectors.
• Innovation, Lean Startup, Design Thinking, UX, user research, business model generation, MVP, pitching and acceleration programs.

UDESC: Colmeia & LabP2D (research)

March 2012 — March 2017

• Developed energy monitoring solutions for data centers (GreenHop project).
• Created algorithms for virtual machine consolidation (CHAVE project).
• Published 13 peer-reviewed research papers in international conferences.

UDESC — Colmeia & LabP2D

2013 – 2016

• Taught introductory and advanced Arduino programming in C++ to undergraduate students.
• Covered embedded systems concepts, sensor integration, and hardware automation.
• Applied in research projects such as GreenHop (data center environmental monitoring).

UDESC — Colmeia & LabP2D

2013 – 2016

• Taught Git fundamentals and collaborative workflows to research lab members.
• Topics included branching strategies, pull requests, and open source contribution practices.

UDESC — Colmeia & LabP2D

2013 – 2016

• Instructed students on LaTeX typesetting for academic papers, theses, and conference submissions.
• Covered BibTeX references, mathematical notation, and IEEE/ACM formatting templates.