Who We Are
How We Work – Our Principles
- Own It. Act like founders, outcomes over titles.
- Transparency, No Ego. Radical candor beats politics; feedback moves us faster.
- Learn Relentlessly. Curiosity is our edge in a world that changes daily.
- Ship Fast, Keep It Simple. Short cycles, minimal complexity, continuous improvement.
- Think Modular. We build with LEGO‑style blocks, ideas, code, products so we can iterate and scale quickly, especially in the AI era.
What Defines Us
- AI Everywhere. Agents run our product and our internal processes. They let us move like a big team, without the overhead.
- High Leverage. Fewer people, bigger impact. No spectators.
- Customer‑Close Experimentation. Tight loops with users let us test, learn, and iterate at speed.
What We Offer
- Ownership & Impact. Your code and ideas reach production and customers fast.
- Rapid Growth. Work in a high-impact team, tackle real complexity, and learn fast through tight feedback loops.
- Ambitious Roadmap. Help craft a foundational platform for global commerce.
- Flexible Hybrid. Milan hub (min. 3 days/week in office) +4 remote weeks each year.
What’s the Role About?
CommerceClarity is building the agent layer for modern commerce — a connected network of task-specialised AI agents that automate catalog, pricing, media and more, compressing weeks of work into minutes. To scale that vision, we’re looking for an engineer who lives and breathes AI: someone who can craft state-of-the-art LLM pipelines, wire them into production, and continuously improve them based on real-world feedback.
Always AI-first, always learning.
This is a hybrid role (70% AI engineering / 30% product) where you’ll work across experimentation, architecture, infrastructure and delivery, all with speed, pragmatism, and ownership.
Key Responsibilities
- Design, build and refine LLM-powered agents using Python (FastAPI, LangGraph/LangChain) and vector databases
- Own the full experimentation loop: prompt/RAG design, offline evals, online A/B tests, post-deployment monitoring
- Develop agent orchestration logic (memory, tool-calling, multi-step planning, latency/cost trade-offs)
- Collaborate with product and data teams to launch new agents and define success metrics
- Champion DevOps for AI: Docker/K8s, GPU orchestration, model versioning, CI/CD for prompts and weights
- Continuously ask: “Could an agent do this?” — and turn the answer into live automation features
Who You Are
Essential:
- 5+ years in software engineering, 2+ years building LLM or NLP systems in production
- Fluency in Italian and English with strong written communication
- Deep Python expertise (async, typing, rigorous testing)
- Hands-on with vector DBs (Weaviate, Pinecone, Qdrant), embeddings and RAG pipelines
- Experience with orchestration frameworks (LangChain, LangGraph, Haystack) and with both open and closed-source models (OpenAI, Mistral, Llama 3, etc.)
- Solid DevOps for AI: Docker, Kubernetes, GPU scheduling, Terraform
- Clear, concise written communication and openness to transparency
Preferred:
- Experience in B2B SaaS, ecommerce, or agent-based workflows
- Contributor to open-source AI-projects or AI infra tooling
- Experience shipping AI-features