AI Engineer

Descrição do trabalho

About Inovretail

Customer satisfaction within retail hinges on the delivery experience, the moment that can make or break the customer’s journey.

Our state-of-the-art order processing and delivery SaaS platform utilizes artificial intelligence and machine learning to transform retail stores into highly efficient picking and packing centers. This enables retailers to provide fast, cost-effective, and dependable collection and delivery solutions to customers, helping to increase sales, satisfaction, and loyalty.

We are dedicated to helping retailers exceed customer expectations by offering greater convenience and flexibility through a variety of reliable and swift order fulfillment solutions. Through Seeplus and our partnerships, including Locky and Uber Direct, we are trailblazing this path, enabling customers to purchase knowing they can receive the goods in a way that suits their needs and schedule.

About the Professional Services Unit

As part of Bright Pixel Capital, the tech investment arm of The Sonae Group, we’re backed by one of Portugal’s largest and most respected business groups. This solid foundation positions Inovretail as a long-term technology partner you can trust.

We’re a SaaS company with a rapidly expanding AI consultancy unit, focused on building smart LLM-powered agents that solve real-world problems. Our team of Data Scientists and Software Engineers works across industries, combining deep R&D expertise with a cutting-edge AI stack.

Our focus areas include:

• Autonomous agents built with LLMs

• Tool-augmented automation (RAG, APIs, databases, planners)

• Scalable, data-driven product solutions

Position Summary

As an AI Engineer, your mission will be to design, build, and deploy intelligent agents that use state-of-the-art LLMs and frameworks. You’ll own full agent lifecycles: from scoping and architecture to deployment and monitoring.

Key Responsibilities

• Agent Design & Architecture
• Define agent types (reactive/proactive/tool-augmented/RAG) based on business needs, leveraging Azure’s event-driven and microservices architecture
• Select appropriate LLMs (GPT-4o, Gemini, Claude, LLaMA, etc.)
• Design orchestrators using LangChain, Semantic Kernel, or custom logic, deployed on Azure Functions or Azure Container Apps
• Architect memory systems (buffer, Redis, MongoDB, summary memory)

• Data & Embedding Pipelines
• Prepare structured (CSVs, databases) and unstructured knowledge (PDFs, FAQs)
• Apply chunking strategies (paragraphs, sliding windows, recursive splitters, etc.)
• Generate embeddings (e.g., text-embedding-3-small, all-MiniLM-L6-v2, etc.)
• Index content into vector stores like Azure Search AI, Pinecone, MongoDB, or FAISS, etc.

• Agent Implementation
• Use tools like LangGraph, Flowise, AgentExecutor, Haystack, etc.
• Integrate APIs, databases, web scraping, and OCR tools.
• Implement planning and tool use strategies (e.g., ReAct, Pydantic planners, etc.)

• Interfaces & Deployment
• Build REST APIs or chat interfaces (Twilio, Slack, etc.) hosted on Azure App Service or Azure Container Apps
• Host on GCP, Azure, or via Docker
• Connect user sessions to memory and file uploads, etc.

• Testing, Monitoring, and Feedback Loops
• Evaluate: hallucinations, relevance, F1-score, etc, using Azure Monitor and Application Insights
• Track user sessions, fallback rates, and latency
• Apply continuous learning from real user logs and feedback

• Qualifications and Competencies
• A Bachelor’s or Master’s degree in Computer Science, Data Science, Artificial Intelligence, Engineering, or a related technical field.
• 3 to 5 years of hands-on experience in AI Engineering, with a proven track record of developing and deploying LLM-powered agents.
• Proficiency in Python, particularly with frameworks such as LangChain, LangGraph, and Semantic Kernel.
• Proficiency in Node.js and TypeScript for backend and API development, and React for building modern, responsive user interfaces.
• Experience with Git for version control and implementing CI/CD workflows using GitHub Actions or similar tools to automate build, test, and deployment processes to Azure services.
• Hands-on experience deploying full-stack applications (React frontend, Node.js/TypeScript backend) to Azure, leveraging Azure’s cloud-native services for secure, scalable, and automated deployments.
• Solid experience working with Large Language Models (LLMs), embeddings, vector databases, and inference pipelines.
• A deep understanding of AI agent architectures, including decision-making flows, tool integration, and memory management.
• Demonstrated