**THIS JOB IS NOT A REMOTE POSITION. ON-SITE WORK IS MANDATORY.**

Snapbau is a digital purchasing and procurement management platform used by more than **320 construction companies in Switzerland**, including major players such as **Orllati, CPSA, Scrasa and ARSA**.

Our platform automates and optimises calls for tender, purchasing workflows, and supplier coordination, helping construction companies **save time, reduce errors, and centralise collaboration** across projects and stakeholders.

You will join the team responsible for **designing and building the applied AI layer** of Snapbau’s digital operating system for procurement and coordination in the construction sector.

This role is hands-on, product-driven, and focused on **real-world automation**, not research.

## Tasks

**AI System Design & Development**

* Design and build **applied AI systems** to automate construction procurement workflows, with a strong focus on:
* Document data extraction (invoices, tenders, contracts, change orders)
* Document classification and tagging
* Entity matching (projects, suppliers, cost codes, budgets)
* Implement **LLM-powered features** for structured document understanding, summaries, and assisted workflows.
* Build and maintain **robust AI pipelines** combining OCR, LLMs, rules, and confidence scoring.
* Develop lightweight APIs and services to integrate AI components into the Snapbau platform.

**Data Engineering & Structuring**

* Work closely with backend developers to:
* Clean, normalize, and structure data from messy or unstructured sources
* Design schemas that support reliable automation and analytics
* Apply construction-specific taxonomies and standards to enable domain-relevant decision-making.
* Improve system accuracy through pragmatic feature engineering and feedback loops.

**Product Integration & Reliability**

* Collaborate with product managers and frontend developers to deliver **AI features that are usable, fast, and trustworthy**.
* Integrate AI systems into Snapbau’s workflows with clear confidence thresholds and fallback mechanisms.
* Monitor performance post-deployment and continuously improve accuracy based on real user feedback.

**Strategy & Evolution**

* Contribute to Snapbau’s **AI architecture and technical roadmap**.
* Evaluate new applied AI capabilities (LLMs, embeddings, agent workflows) and assess their real value for construction tech.
* Help shape internal best practices around AI tooling, evaluation, and deployment as the team scales.

## Requirements

* Proven experience building and deploying **applied AI or ML systems** in production environments.
* Strong programming skills in **Python**; solid working knowledge of **SQL**.
* Hands-on experience with:
* NLP or document processing
* Classification, matching, or recommendation systems
* LLM APIs (OpenAI, Anthropic, or similar)
* Experience integrating AI systems via **REST APIs** into real products.
* Strong ability to work with **messy, unstructured data** and turn it into reliable outputs.
* Familiarity with ML frameworks such as **PyTorch, TensorFlow, scikit-learn**, or equivalent.
* Experience working in fast-paced, cross-functional product teams.
* Willingness to learn construction-specific standards (prior experience is a strong plus).

## Benefits

Join Snapbau if you want to **build AI that actually runs the real economy**.

You’ll work on problems that matter, with:

* High technical ownership
* Direct impact on hundreds of construction companies
* A fast path to shaping how AI is applied in one of the most complex and underserved industries

**This role is:**

* Product-focused
* Accuracy-obsessed
* Grounded in real business constraints
* About building systems that work in production

**This role is not:**

* Academic research
* Model training for its own sake
* A “prompt-only” position
* Fully remote