Software Architect (m/w/d)

Beam AI Inc.; Berlin; On-site; EUR 140,000 - 200,000; Full time

Job summary

Beam AI in Berlin seeks a seasoned Software Architect to lead the design of GenAI and agentic systems. You will define the architectural north star for building intelligent, autonomous agents that reason, learn, and operate at scale.

You will shape orchestration frameworks, model experimentation pipelines, and ensure engineering choices align with product goals and cost efficiency, while mentoring senior engineers.

Qualifications

• 10+ years of software engineering experience, with 3+ years architecting AI/ML systems at scale.
• Expertise in LLMs, agentic frameworks, RAG, vector databases, and orchestration layers.
• Strong backend skills in Node.js, Python, or Go.
• Cloud platforms (AWS, GCP, or Azure) and MLOps tooling.
• Proven track record aligning architecture with business strategy.

Responsibilities

• Define the architecture vision for GenAI and agentic systems, including orchestration, memory, and reasoning layers
• Design modular, scalable systems that support LLMs, RAG pipelines, vector databases, and multimodal inputs
• Architect agentic workflows capable of multi-step reasoning, autonomous decision-making, and long-context operations
• Lead technical decision-making across key systems, ensuring alignment with product goals and cost efficiency
• Build frameworks for experimentation, model evaluation, and fast iteration in production
• Partner with product, ML, and engineering to bring new features to life, with speed and strategic clarity
• Drive scalability, observability, and reliability across inference, retrieval, and orchestration pipelines
• Translate architecture into business impact: act as a technical voice in leadership conversations and client engagements
• Mentor senior engineers, shape design standards, and cultivate architectural excellence across the team

Skills

LLMs

Agentic frameworks

RAG pipelines

Vector databases

Orchestration layers

Node.js

Python

Go

Cloud platforms

Architectural leadership

Education

Bachelor's degree in CS or related field

Tools

LangChain

LangGraph

LlamaIndex

OpenAI API

Anthropic API

Mistral

Job description

Beam AI builds self-learning AI agents and an organisational operating system where intelligence grows through autonomous discovery. We help enterprises evolve from AI-curious to AI-native with automation that delivers results today and learns to improve itself over time. Join us in shaping a future where AI continuously discovers, adapts, and makes work more meaningful. To learn more, visit https://beam.ai/.

About The Team

Beam's engineering team is building the core infrastructure for self-learning agents, turning AI into reliable, production-grade systems. We work across the full stack, from frontend to infra, solving for real-time orchestration, tool integration, and complex workflow execution. This is a high-trust, high-impact environment where engineers move fast, ship often, and help shape both the product and the platform as we scale. If you care about building with clarity, speed, and depth, you'll do your best work here.

About The Role

We're looking for a seasoned Software Architect to lead the design of our GenAI and agentic systems. You'll define the architectural north star for how Beam builds intelligent, autonomous agents that reason, learn, and operate at scale. This is a high-impact role for someone who can bridge visionary product thinking with deep technical execution. You'll influence everything from our orchestration frameworks to model experimentation pipelines, and ensure our technical choices align with strategic outcomes.

Responsibilities

• Define the architecture vision for GenAI and agentic systems, including orchestration, memory, and reasoning layers
• Design modular, scalable systems that support LLMs, RAG pipelines, vector databases, and multimodal inputs
• Architect agentic workflows capable of multi-step reasoning, autonomous decision-making, and long-context operations
• Lead technical decision-making across key systems, ensuring alignment with product goals and cost efficiency
• Build frameworks for experimentation, model evaluation, and fast iteration in production
• Partner with product, ML, and engineering to bring new features to life, with speed and strategic clarity
• Drive scalability, observability, and reliability across inference, retrieval, and orchestration pipelines
• Translate architecture into business impact: act as a technical voice in leadership conversations and client engagements
• Mentor senior engineers, shape design standards, and cultivate architectural excellence across the team

Requirements

• 10+ years of experience in software engineering, with 3+ years architecting AI/ML systems at scale
• Expertise in LLMs, agentic frameworks, RAG, vector databases, and orchestration layers
• Strong backend engineering skills in Node.js, Python, or Go
• Deep understanding of cloud platforms (AWS, GCP, or Azure) and MLOps tooling
• Track record of aligning architecture with business strategy and measurable outcomes
• Exceptional communication skills across technical and executive audiences
• Proven mentorship and technical leadership experience

Nice to Have

• Experience with LangChain, LangGraph, LlamaIndex, or similar orchestration frameworks
• Familiarity with open-source model stacks (Llama, Mistral) or commercial APIs (OpenAI, Anthropic)
• Exposure to multi-agent systems and long-running autonomous workflows

Benefits

At Beam, we're building an environment where ambitious people can do their best work, with clarity, purpose, and room to grow. We're thoughtful about what we build, how we work, and who we hire. The problems we're solving are real. The systems we're building are complex. And the people here care deeply, about craft, about speed, and about doing work that actually matters. You won't find rigid hierarchies or endless process here. You'll find high trust, high standards, and a team driven by curiosity, ownership, and long-term thinking.

Our culture is shaped by our values in action

AI-native thinking: Every teammate thinks in agentic workflows, uses AI tools daily, and looks for leverage through automation. AI isn't just what we build, it's how we operateCustomer obsession: We build with urgency and empathy, speak to customers early and often, and measure success through their outcomesSpeed as a habit: We ship fast, learn faster, and prioritize unblocking over perfection. We favour small experiments with tight feedback loopsLeverage-focused execution: We hire high-impact players, invest in the 20% that drives 80% of results, and value candid feedbackHighly aligned, loosely coupled: We align on outcomes, not tasks. You'll work independently, but never in isolationData-driven, human-guided: We seek clarity over comfort, speak up early, and use feedback to grow together