Petro – LangChain, Python, RAG, experts in Lemon.io

Petro

From Portugal (UTC+1)flag

AI Engineer|Middle-to-senior

Petro – LangChain, Python, RAG

Petro is an AI engineer with 8 years of full-stack experience in TypeScript, Node.js, and React, and he builds AI services in Python. For the past 2.5 years, he has owned the AI features of a live-learning EdTech platform, shipping five of them to production. Petro works with LangChain, LangGraph, RAG, and pgvector, and designs for production: he cut projected LLM costs by 35–40% and put guardrails in place for tenant data. He has led and coached engineering teams and works directly with CTOs and VPs of Engineering to turn early ideas into features the team can build.

11 years of commercial experience in
Asset management
E-commerce
Edtech
Retail
CRM
SaaS
Main technologies
LangChain
1 year
Python
2.5 years
RAG
1 year
LangGraph
1 year
Vector Databases
1 year
Direct hire
Possible
Ready to get matched with vetted developers fast?
Let’s get started today!

Experience Highlights

Technical Lead
May 2024 - Jul 20262 years 2 months
Project Overview

An EdTech platform for real-time, instructor-led classes with AI-driven learning features, built for active learning rather than general video conferencing. Its course-generation module turns an uploaded video into a complete lesson.

Responsibilities:
  • Led the development team and architected the AI systems for the platform, coordinating AI feature delivery across the live classes, lesson modules, and infrastructure teams.
  • Delivered AI-generated chat summaries and live quizzes for real-time instructor-led sessions with 100–1,000 learners.
  • Extended the video processing pipeline with AI workflows that generate ready-to-use lessons and learning activities with measurable learning objectives.
  • Reduced projected LLM costs by 35–40% by routing tasks to lower-cost models and applying prompt engineering.
  • Redesigned the AI services architecture into modular services that are switched on per tenant according to their payment plan.
  • Co-developed an AI agent for smart search across lesson materials, after repositioning an AI assistant concept as smart search over the original lesson content.
  • Turned leadership ideas into actionable requirements for seven AI-driven features, analyzed competing EdTech platforms to identify three new types of AI-generated learning activities, and prepared user-focused use cases for the design team.
Project Tech stack:
Typescript
Python
FastAPI
React
LangChain
OpenAI
Gemini API
AWS
Terraform
Docker
RAG
AI agent development
Prompt engineering
Kubernetes
LangGraph
Lead Full-stack Engineer
Nov 2020 - May 20243 years 6 months
Project Overview

A development company where I worked on several client products: a Shopify returns app that resells returned items through local partner stores to cut merchants’ logistics costs, a restaurant inventory management system connecting restaurants with suppliers, and more.

Responsibilities:
  • Owned architecture and feature delivery for several MVPs and core platform features, defining MVP requirements that balanced long-term extensibility with lean delivery.
  • Initiated and led the Shopify returns app MVP, owning its architecture, infrastructure, CI/CD, and feature delivery.
  • Led the migration from Stripe to Adyen, reducing payment-processing fees and increasing marketplace host revenue.
  • Cut local and CI test execution time by 10x by migrating Jest transforms to SWC.
  • Replaced XLS-based form configuration with a form builder supporting dynamic fields, validations, grid layouts, and persistent ordering.
  • Introduced Terraform infrastructure as code as a reusable package for provisioning new client environments.
  • Led and coached a team of 4 front-end engineers and helped them grow into full-stack developers. Redesigned the hiring process for new engineers, which improved probation success rates and team culture fit.
  • Worked closely with clients, product owners, and project managers to define, estimate, and ship user-facing features, and mentored teammates across projects while improving team documentation.
Project Tech stack:
Node.js
Typescript
SQL
React
Preact
CI
CD
AWS
GCP
Full-stack Developer
Jul 2017 - Jan 20191 year 6 months
Project Overview

A no-code platform for building bots and workflows, with chatbot features and platform integrations.

Responsibilities:
  • Built and shipped chatbot features and platform integrations across the frontend and backend.
  • Built 8 configurable Lambda-based workflow components for the no-code bot and workflow platform.
  • Created an onboarding flow for new engineers that reduced onboarding time.
  • Analyzed product requirements and worked closely with QA and UX to ensure high-quality implementations.
Project Tech stack:
AWS Lambda
Vue.js
Node.js
Front-end Developer
Apr 2016 - Aug 20171 year 4 months
Project Overview

A global digital consultancy, systems integrator, and e-commerce agency. Petro built storefront features for clients’ e-commerce sites on the Hybris platform.

Responsibilities:
  • Built and customized storefront features on Hybris, including advanced product filtering, product detail pages, and product listing pages, using jQuery and Vue.js.
  • Solved cart state sharing on a storefront that mixed HTTP and HTTPS pages by using a hidden HTTPS iframe as a shared store, keeping cart data consistent across secure and non-secure pages.
  • Worked with the team to define features, turning loosely specified client requests into front-end solutions that could be built on legacy enterprise platforms.
Project Tech stack:
jQuery
Vue.js
JavaScript
Java
SCSS
Webpack
Gulp

Education

2016
Computational Mathematics
Bachelor's degree

Languages

English
Advanced

Hire Petro or someone with similar qualifications in days
All developers are ready for interview and are are just waiting for your requestdream dev illustration
Copyright © 2026 lemon.io. All rights reserved.