Original listing text, shown exactly as published by the company.
🎯 What you will do
As an AI Software Engineer, you will own the development and continuous improvement of Joko AI — our AI shopping agent that seamlessly guides users to the ideal product whatever their intent — helping them save significant time by taking into account all relevant considerations (price, quality, reviews, environmental impact), across every product sold by any e-commerce merchant in the world. Working closely with Alex, our CTO, and alongside a small team of world-class engineers, sitting at the intersection of applied research and production.
- Agent architecture: You will own the architecture of our agentic system — ensuring it constantly evolves, incorporates the latest advancements in the field, challenges existing assumptions, tests bold ideas, and pushes the boundaries of what our AI assistant can do.
- Evaluation & monitoring: You will design and run robust evaluation frameworks to measure Joko AI's performance, catch regressions, and build confidence before shipping to production.
- Capabilities & integrations: You will constantly expand the set of tools and capabilities that Joko AI can leverage — web search, vector search, ever-smarter catalog lookups — making the agent more resourceful, faster, and more reliable at scale.
- Catalog & data at scale: Great recommendations start with great data. Joko AI is powered by a product catalog of hundreds of millions of entries, continuously updated. You will tackle the massive challenges of ingesting, standardizing, and enriching this data at scale, where AI itself plays a key role.
- Personalization: Joko has access to uniquely rich data — bank transactions and browsing behavior through the browser extensions — giving us unparalleled insight into user preferences and shopping patterns. You will leverage these signals to deeply understand user intent and build a truly personalized experience — making Joko AI feel like it genuinely knows what you're looking for.
- Experimentation & prototyping: You will constantly push Joko AI forward by identifying new challenges and building new features for the agent. You will prototype fast, put early versions in users' hands, and iterate based on real feedback
- Staying ahead: You will stay on top of the latest models, agentic frameworks, and research papers, and translate what's emerging into concrete improvements for Joko AI.
- Fine-tuning & optimization: You will fine-tune models to improve quality, reduce latency, and cut costs — always with a production-grade mindset.
- Runtime & scalability: You will iterate on the backbone and infrastructure Joko AI runs on, ensuring it handles hundreds of thousands of daily users with zero downtime and minimal latency.
- Product collaboration: We are inventing an entirely new shopping experience with Joko AI — and much of what the ideal UX looks like is still to be defined. You will work hand-in-hand with the Product team, exploring bold design directions, imagining what the future of AI-assisted shopping can be, and turning that vision into reality from early concept to full rollout.
👀 Who we're looking for
- Experience: You have 5+ years of software engineering experience, including time building and shipping production systems. You've operated at a senior or staff level and you're used to owning ambitious projects from design to production.
- Track record: You've ideally built production AI systems that real users rely on, and you care deeply about code quality, separation of concerns, and testing.
- Technical skills: You are deeply familiar with agentic frameworks (LangChain, LangGraph, or equivalents), LLM observability tools (LangFuse or equivalents), LLM APIs, and evaluation tooling for AI systems. An AI/ML background or degree is a strong plus, but not a blocker.
- Builder's mindset: You love building from scratch. You are not precious about your ideas — you ship, test, learn, and iterate fast.
- Full-stack spirit: You are comfortable going from AI logic to infrastructure to frontend prototyping when needed. You don't stop at your lane.
- Fast learner: The AI landscape moves fast. You thrive in it. You follow the research, experiment with new models and tools, and you are always slightly ahead of the curve.
- Problem solver: You approach hard technical problems methodically. Latency, cost, hallucinations, long-term memory — you know these challenges and you enjoy tackling them.
- Team player: You work closely with Product, iterate with users, and communicate clearly about what's working and what isn't.
- Drive & ownership: You take full ownership of your scope. You don't wait to be asked — you spot issues, propose solutions, and ship.
- Language: You are fluent in English (written and spoken).
- Mindset: You have an entrepreneurial mindset, you like challenges, you welcome feedback and you love constantly learning and getting better.
