Founding AI Agent Engineer

Allus AI is building the next generation of vision intelligence systems.

We are building an AI-native platform for real-world vision workflows, combining vision intelligence and modern agentic software systems. Our goal is to make it dramatically easier for users to go from a problem statement and example data to a reliable production workflow.

We’re hiring a Founding Engineer to help build the core software systems powering the platform. This role sits at the intersection of agentic workflows, product engineering, developer infrastructure, and AI-native user experience. We’re backed by leading investors including Y Combinator, Pioneer Fund, Blast, Transpose, 007 Venture Partners, K50, and Ponderosa.

What you’ll do

• Own the architecture and development of core AI-native software systems across the platform
• Build agentic workflows that turn user requirements and example inputs into structured, multi-step product workflows
• Design systems for orchestration, tool use, task planning, state management, and workflow reliability
• Build software that supports data intake, review loops, human feedback, and iterative task refinement
• Develop infrastructure for prompt orchestration, memory, retrieval, and contextual personalization
• Build internal tools and user-facing product features powered by LLMs and agents
• Design evaluation and monitoring systems for workflow quality, reliability, latency, and production readiness
• Improve developer velocity through strong abstractions, automation, and AI-native tooling
• Rapidly prototype and ship new AI-driven product capabilities

What we’re looking for

• 5+ years of experience building production software systems
• Strong backend engineering skills, including APIs, data pipelines, and system design
• Experience building and shipping LLM-powered applications or agentic product workflows
• Strong understanding of prompt design, tool use, retrieval, orchestration, and evaluation techniques
• Ability to move quickly from prototype to production
• Strong product intuition and desire to build user-facing AI experiences
• Comfort working across multiple parts of the stack with high ownership

Relevant experience

• Full-stack software development
• Agent orchestration systems and multi-step workflow design
• RAG systems, embeddings pipelines, or retrieval architectures
• Human-in-the-loop systems and review workflows
• Evaluation systems, observability, and production monitoring
• Internal platforms, developer tooling, and automation systems

AI-native engineering
We are building an AI-native company and expect engineers to actively use modern AI tools in their daily workflow. Strong candidates should already be comfortable using tools such as Cursor, Claude Code, modern AI coding assistants, and LLM-based and agent-based development workflows.

Technologies we use
Python, Rust, SQL, TypeScript, React, Node.js, FastAPI, LangChain, Docker, Redis, AWS, GCP.

1. Intro Call

This is an informal conversation to learn more about your background, interests, and what you are looking for. We will also share more about Allus, the role, and what it means to join as a founding engineer.

We are especially interested in candidates who enjoy startups, high ownership, fast execution, and building in ambiguous environments.

2. Technical Screen

This is a practical 60-minute coding interview focused on the kind of systems we actually build.

Depending on your background, this may include APIs, data models, workflow orchestration, retrieval, internal tools, or LLM-powered product features. We care much more about how you build useful systems than about algorithm puzzles.

You are welcome to use your preferred development environment and AI tools such as Cursor, Claude Code, or ChatGPT. AI-native engineering is part of how we work at Allus.

3. Founder / Final Interviews

This stage is a deeper evaluation of your technical ability, product thinking, communication, and overall fit.

We will likely discuss your past work, ownership, and how you operate in fast-moving environments. We may also go deeper on systems you have built, how you make technical and product decisions, and how you think about speed, quality, and shipping to real users.

Depending on the role and your background, this round may include a mix of:

• product and systems design
• building or extending a realistic feature
• debugging and iteration
• discussion of product judgment and founding-team fit

We tailor this stage to the candidate and the role.

What We Evaluate

Across the process, we are generally looking for:

• strong software engineering fundamentals
• backend or full-stack product judgment
• ability to ship production systems
• comfort with ambiguity and ownership
• experience with LLMs, agents, retrieval, or workflow systems
• strong communication and pragmatic decision-making