In 2021, Ingram Technologies partnered with Siguelaola to build Ola, a fintech prototype focused on financial literacy and accessible trading in Latin America. This project involved integrating machine learning into trading workflows, developing an AI-powered robo advisor, and contributing to open-source fintech tooling.
Siguelaola wanted an investment app that was easy to use and powered by AI. The reference point was: commission-free trading, no account minimums, fractional shares. We implemented intelligent automation and deeper analytics. The technical goals included a clean UI, fast trading execution, strong portfolio tracking, and AI-driven guidance.
We used a hybrid-microservices architecture to support modular deployment, fault isolation, and scalability without making the stack too complex, thanks to FastAPI and NextJS, with some AWS Lambda services on top.
We implemented ML models for price prediction (regression), market direction (classification), and trading strategy optimization (deep reinforcement learning). NLP models processed financial news and social media for sentiment analysis. These tools helped uncover market patterns at speed and scale. ML enabled more accurate market forecasts, faster signal detection, and the automation of trading strategies. We also accounted for model overfitting, rare events, and the need for frequent retraining.
Ola’s robo advisor gathered user data (financial goals, risk profile) and generated portfolio recommendations. It offered automatic fund allocation, portfolio rebalancing, and potentially tax-loss harvesting. The goal was simple: offer users a low-cost, algorithm-driven alternative to traditional investment advisors.
Regulatory Compliance: Our work was thoroughly reviewed and audited by a third-party security firm. Thanks to this, Siguelaola obtained broker-dealer registration, ensuring legal compliance and user protections. This required satisfying operational and financial oversight standards. It also added trust for users and partners.
Alpaca offers strong APIs for trading apps. We partnered very early and assisted in uncovering several bugs and issues with the platform. At the time, there was no SDK for Alpaca. We spearheaded the effort in developing this SDK, which was eventually acquired by Alpaca and made open source as a collaboration. This SDK is now the public Alpaca Python SDK and has since been significantly expanded.
To meet KYC/KYB compliance requirements, together with Siguelaola we collaborated on building a no-code platform for KYC/KYB analysis and validation. This work was created with the added partnership of Moneygram and was dubbed "Ossky" — an open-source suite of onboarding tools. Moneygram also contributed regulatory and verification expertise. Ossky automated document and identity verification, beneficial ownership checks, and risk scoring. It helped fintech apps meet AML/CTF requirements while speeding up user onboarding.
Ola was selected for the inaugural Techstars Payments cohort in 2022, recognized for its identity orchestration capabilities. The program offered access to mentors and partners, validating the technical and product approach.
In this project, we combined modern architecture, applied machine learning, and open-source tools to deliver a working prototype of an intelligent fintech app. The project highlighted how AI can improve financial services, from user onboarding to portfolio management. It also reinforced the value of open collaboration and tight integration between technology and compliance. This deep partnership lasted a total of two years and was a real success.
© Ingram Technologies
Ingram is an EU-based Artificial Intelligence R&D lab established in 2022.
We work on ethical, compliant and open AI-based software solutions for startups and SMEs.