Artificial Intelligence Engineer
Pubblicato il 04-10-2026 - Duferco in Milano
We are expanding Duferco's Corporate Innovation Team: a fast-moving engineering team operating in a startup-style environment, backed by the stability and scale of a global diversified group.
Possiede le competenze e l'esperienza giuste per questo ruolo?
Continui a leggere per scoprirlo e invii la sua candidatura.
You will work on a wide variety of projects across domains such as energy, trading, logistics, steel, insurance, sports, and other sectors — not as an external consultant, but as part of the group itself, with the ownership and continuity of an internal product and engineering team.
You will build AI-enabled products, agents, automation systems, evaluation pipelines, and backend infrastructure that remove operational bottlenecks, improve decision-making, and create leverage across the business.
This is a great role for an engineer who is excited about applying modern AI systems to real business problems, cares about shipping reliable products, and wants to grow quickly alongside experienced teammates.
About the Team
We sit close to real business problems and have the freedom to prototype and ship quickly.
AI is already part of our daily workflow, and we expect that role to keep expanding.
We care about building things well, shipping systems that create leverage, and working with people who make the team better.
We appreciate engineers who enjoy ownership, fast iteration, pragmatic problem-solving, and AI-assisted development.
What You'll Do
Design and build AI-powered systems that make the organization faster and smarter:
LLM-powered applications, agents, assistants, and automation workflows
Retrieval-augmented generation systems, internal knowledge tools,
and document intelligence pipelines
Evaluation, monitoring, and feedback loops for AI systems in production
Backend services and APIs that connect AI capabilities to real business workflows
Integrations across internal and third-party systems, often where the hard part is understanding the domain, not the model
Data pipelines and tooling that support experimentation, deployment, and continuous improvement
Engineering tooling that improves how we build, test, and ship AI-enabled products
You will work with more experienced engineers and domain experts to understand business problems, design pragmatic AI solutions, and build, test, ship, evaluate, and improve systems over time.
You will not just experiment in notebooks.
You will be encouraged to understand the domain, ask hard questions, suggest improvements, and take on increasing ownership of production systems.
What We Offer
A role with real ownership, support, and room to grow
Exposure to a broad variety of technical and business problems
The opportunity to build practical AI systems used inside a diversified international group
Modern AI-assisted engineering tools and time to explore new technologies and approaches
A fast-moving,
collaborative environment with talented and ambitious people
Competitive compensation and benefits
Flexible working hours and a remote-first setup, with hybrid options where applicable.
You're Probably a Fit If
You have built or contributed to LLM-powered applications, agents, RAG systems, automation tools, or similar AI-enabled products
You are comfortable working with APIs, databases, backend services, and modern cloud-based infrastructure
You understand that building useful AI systems is not just about prompting models, but also about product thinking, data, evaluation, reliability, and integration
You can take a messy business problem and turn it into a practical technical solutionYou make pragmatic trade-offs between speed, quality, cost, latency, and maintainability, and can explain why
You are curious about how models behave in real workflows and care about measuring whether the system actually works
You use AI coding tools daily — Claude Code, Codex, Cursor, or similar — and have opinions about where they help and where they don't
You enjoy working close to the business and take the time to understand a process before modeling or automating it
Experience with LLM APIs, open-source models, vector databases, embeddings, evaluation frameworks, agent frameworks, MCP servers, data warehouses such as Redshift, Postgres, FastAPI, or AWS-based services is a strong plus.
xysqume
You do not need experience with every technology we use.
Strong engineering ability, adaptability, product judgment, and willingness to learn matter much more.
If this sounds like the right fit for you, we'd love to hear from you.
#J-*****-Ljbffr