AI Engineer

Pillar; Lombardia; On-site; EUR 55,000 - 75,000; Full time

Job summary

Pillar is building the financial operating system for construction, using AI and modern software. We seek an AI Engineer who can ship production-grade AI features, from problem framing to deployment, with a product-minded, startup-driven approach.

The role emphasizes end-to-end ownership, agentic workflows, and AI-native product integration across a TS stack. Location: Milan or Barcelona, English working proficiency required.

Qualifications

• Experience building LLM-based applications in production.
• Hands-on with AI orchestration, agents, structured outputs.
• Strong background in embeddings, RAG, vector databases.
• Proficient in SQL and relational databases.

Responsibilities

• Own AI features end to end, from problem to production.
• Design and build agentic workflows with tool orchestration and multi-step reasoning.
• Build RAG and semantic search pipelines using embeddings and vector databases.
• Develop document extraction workflows turning invoices and forms into structured data.
• Establish observability, evaluation, tracing and quality checks for AI features.
• Improve retrieval and prompting strategies to boost accuracy and product quality.
• Contribute across the TypeScript stack for product integration.
• Move quickly from ambiguity to working product with pragmatic decisions.

Job description

Pillar was founded in 2025. Today, we are 80+ people, serve 900+ customers, raised €12 million in 2026, and are expanding internationally.

We are building the financial operating system for construction: bringing AI, automation, and modern software to an enormous industry still full of slow, fragmented processes.

We are looking for an AI Engineer who knows how to turn LLMs into reliable product experiences, moves fast, and wants to build.

The mission

Build production-grade AI systems that solve real customer problems.

We want someone who can move beyond prototypes and build agents, retrieval systems, document pipelines, and AI-native product features that are reliable, measurable, and deeply integrated into Pillar.

What you will do

• Own AI features end to end, from product problem to production.
• Design and build agentic workflows using tool orchestration, structured outputs, and multi-step reasoning.
• Build RAG and semantic search pipelines using embeddings, vector databases, hybrid search, and re-ranking.
• Develop document extraction workflows that turn invoices, reports, and forms into structured data.
• Build observability and evaluation systems for AI features, including tracing, experimentation, and automated quality checks.
• Optimize context, retrieval, and prompting strategies to improve accuracy and product quality.
• Work across our TypeScript stack when AI features require it, from backend logic to product integration.
• Use AI as a native part of your own engineering workflow, not only as something you build for customers.
• Move quickly from ambiguity to working product, making pragmatic technical decisions along the way.

Our stack

• AI: Mastra · LLM APIs · embeddings · vector search · AI-native workflows

Who we are looking for

AI engineers with product mindset and high agency

• You have experience building LLM-based applications in production, not only prototypes.
• You are comfortable with agents, tool use, structured outputs, and orchestration.
• You have hands-on experience with RAG, embeddings, vector databases, and retrieval strategies.
• You are comfortable with TypeScript or can become productive in it quickly.
• You have worked with AI orchestration frameworks such as Mastra, LangChain, CrewAI, or similar.
• You are strong with SQL and relational databases.
• You understand prompt design, evaluation, and the trade-offs involved in building reliable AI systems.
• You have worked in a startup or scaleup environment and understand the pace, ambiguity, and ownership it requires.
• You take initiative, get your hands dirty, and get things done without waiting for perfect instructions.
• You can zoom out to understand the full product problem and zoom in when technical details matter.
• You communicate openly, learn quickly, and collaborate without ego.

How we work

• Agile and squad-based: small teams own their domains, with no unnecessary layers.
• AI-first: we build and work in the age of AI.
• Product ownership: engineers are hands-on and accountable for delivered value.
• Bottom-up: initiative is expected, not delegated.

The profile that can succeed

AI-native · Product-minded · Startup DNA

The sweet spot: a hands-on engineer who has already built real AI products, ideally in a startup or scaleup environment, and knows how to combine LLM capabilities, software engineering, data, and product judgment to ship reliable features.

Location & languages

Milan or Barcelona; professional working English.

Hiring process

01: CV screening.

02: 30-min Talent Call - motivation, Pillar knowledge, mindset, and startup fit.

03: Two 30-min Technical Interviews - one focused on AI systems and architecture, one focused on product engineering and execution.

04: Final Interview with our CTO.

What we offer

• Gross salary 55,000-75,000 € depending on experience.
• The opportunity to build core parts of Pillar’s AI product and engineering foundations as we scale.
• End-to-end ownership: from AI architecture and experimentation to shipping features used by real customers.
• The chance to work in an AI-native engineering team, where AI is part of both the product and the way we build it.
• Close collaboration with Product, AI, and Engineering, with direct exposure to the problems we are solving for customers.
• Real autonomy, responsibility, and room to grow as Pillar and the engineering team scale.