About us
Regulatory submissions that take months. Deviation reports that pile up on QA desks. Pharmaceutical teams spending their best hours os documentation instead of science.
We are building the AI systems that change this.
Visium is an enterprise AI company with deep roots in life sciences. Our two products, Scribe and Devin, are purpose-built for pharma teams operating in GxP-regulated environments, and are gaining traction with leading life sciences organizations across Europe. Scribe automates the drafting of regulatory documents, keeping human experts in control at every step. Devin handles deviation management end to end, from event capture to root cause analysis and corrective action generation.
We are now building the team that takes these products further. If you want to work on AI that genuinely matters, this is the role.
Role
You will own and evolve the agentic layer that powers our products. That means designing and building the multi-agent orchestration systems, RAG pipelines, tool-use architectures, and workflow engines that sit at the core of what we build.
This is a senior engineering role with real production stakes. You will make foundational architecture decisions, shape engineering standards, and work directly with product and leadership from day one. It is not a research role. It is a building role.
Key Responsibilities:
- Design and implement multi-agent architectures using Agno, LangChain, and LangGraph
- Build and maintain scalable backend services and APIs in Python and FastAPI
- Own agent orchestration logic: task routing, tool use, state management, and human-in-the-loop integration
- Design and implement RAG pipelines over domain-specific data, with strict requirements for accuracy, traceability, and compliance
- Make production-grade decisions on LLM integration, prompting strategies, and reliability patterns
- Set engineering standards for code quality, testing, and deployment of agentic systems
- Drive end-to-end feature ownership from design through to production
- Mentor engineers and actively shape the engineering culture as the team grows
- Collaborate closely with product and leadership to influence the technical roadmap
Requirements
You bring technical strength, curiosity, and a drive to build meaningful AI systems. Here is what helps you succeed in this role:
- A strong Python engineer who has shipped production systems, not just prototypes
- Hands-on with agentic frameworks (LangChain, LangGraph, Agno, or similar)
- Fluent in LLMs in production: prompting, tool use, orchestration, failure modes
- Experience building production RAG systems: retrieval architectures, vector databases, embedding strategies, and evaluation frameworks
- Experience with agent evaluation frameworks and observability tooling (e.g. LangSmith or similar): tracing agent behavior, debugging multi-agent interactions, and ensuring reliability in production
- Comfortable with cloud environments (AWS, GCP, or Azure) and distributed system design
- Experienced with MLOps practices: CI/CD, monitoring, deployment pipelines
- Autonomous, low ego, and serious about what you build
- Excited about working in a domain where the quality of your work has real-world consequences
Nice to have:
- Experience in life sciences, regulated industries, or GxP-compliant environments
- Experience with multi-agent systems and workflow orchestration
We know that great candidates might not check every box. If you’re excited about this role and believe you could contribute, we’d love to hear from you.
Benefits
What we offer
- A competitive compensation package
- A flexible working culture because your work-life balance matters to us
- A position that enables you to have an impact on 1’000s of people, and the whole company's growth.
- An international, knowledgeable, and passionate team with a strong collaborative mindset
- Employee Stock Ownership Plan
Why Join Us
- Work with autonomous agents, orchestration runtimes, and knowledge retrieval systems
- Build technology that transforms how life-science experts research, decide, and create
- Shape architecture, culture, and direction from the very beginning
- Collaborate with world-class engineers and researchers who share a deep passion for AI
- Hybrid setup with high trust, autonomy, and flexibility
- Competitive compensation and the chance to be part of a company defining a new AI category
Skills Required
- Strong Python experience with production systems
- Hands-on with agentic frameworks (LangChain, LangGraph, Agno)
- Fluent in LLMs in production
- Experience building production RAG systems
- Experience with agent evaluation frameworks
- Comfortable with cloud environments (AWS, GCP, Azure)
- Experienced with MLOps practices
- Experience in life sciences or regulated industries
- Experience with multi-agent systems
What We Do
Visium is a leading Swiss AI & Data consultancy helping enterprises unlock the full business value of AI. We combine deep technical expertise with a strong consulting foundation to design and deliver strategic, scalable, and compliant AI solutions. We've delivered over 150 AI solutions for some of the world’s leading enterprise, helping them drive efficiency, unlock new revenue streams, and scale solutions that deliver long-term value. From strategy to deployment, our mission remains the same – to build AI that drives measurable, responsible impact at scale. We believe in using AI as a force for good. Through our Visium4Good initiative, we support NGOs and social impact projects on a pro bono basis, using AI to drive positive change where it’s needed most.
