What the role involves
About Yuma AI
- Yuma is building a next-generation orchestration platform that deploys autonomous AI agents dedicated to customer support in e-commerce.
- We already support 150+ paying merchants and help them automate up to 80% of their customer tickets freeing human agents for higher-value conversations.
- Founded by Guillaume Luccisano (3rd-time YC founder) at the end of 2022, Yuma is one of the leaders in the AI Support Automation space with 10+ consecutive months of double-digit growth.
- We are now a team of 25 passionate and ambitious people based between Paris, Barcelona and Boston and have grown revenue 5x in 2025. We aim to repeat this trajectory in 2026. The scale-up phase is fully underway. 🚀
About the Role
This is a foundational role. You will be joining Yuma as one of the core engineers on a tight, high-performing team of 5. We move fast, iterate aggressively, and care deeply about product quality and user value.
Over the past two years, we’ve built an impressive amount of product surface area, AI systems, and automation. Yet what lies ahead is even bigger. As we scale our platform, models, and AI agents, your work will have a direct and immediate impact on the experience of thousands of users.
This role goes far beyond classical backend development. You will work across the full product lifecycle: new features, A/B tests, model integrations, AI agent pipelines, and performance optimization. If you enjoy working at the intersection of backend engineering, product iteration, and applied AI, you will thrive here.
Backend Engineering
- Architect, build, and scale large Ruby on Rails applications
- Own major systems within our Rails API, including performance, reliability, and maintainability
- Manage PostgreSQL at scale: schema design, indexing, performance diagnostics
- Maintain and optimize Redis + Sidekiq pipelines for high-throughput workloads
- Lead backend-heavy feature projects, from design to deployment
AI / LLM Systems
- Integrate, evaluate, and iterate on LLM providers and models
- Design systems that manage non-deterministic model outputs safely and predictably
- Build and own A/B tests, eval pipelines, and agentic workflows
- Optimize for latency, reliability, and cost across LLM-heavy features
- Contribute directly to prompts, agents, and AI system architecture
Product & Full-Stack Contribution
- Collaborate tightly with founders and engineers to define product requirements
- Jump into frontend (Next.js/React) when needed to ship features end-to-end
- Run rapid product iterations: fast feedback loops, experimentation, data-driven decisions
- Maintain a high bar for product quality while optimizing for speed of execution
The Ideal You
- You have 5–10+ years of experience building and scaling production Rails applications
- Rails has no secrets for you : you know how to ship quickly and safely
You are pragmatic
- you make thoughtful trade-offs that maximize user value and iteration speed
- You naturally take ownership and thrive in high-velocity environments
- You have recent, hands-on experience with LLMs and already use AI in your day-to-day workflow
- You’re comfortable navigating ambiguity, experimenting, and working with non-deterministic systems
- You can contribute across the stack when needed (backend-first, but not backend-only)
- You deeply care about product quality, speed, and delivering a great user experience
The Alternative You
If you're still early in your career but see yourself becoming the engineer described above within the next 2–3 years, don’t hesitate. We consistently leave room for raw talent and high potential, especially for people who learn fast, care deeply, and want to grow.
Why Yuma?
- High impact with ownership from day one : join a small, international engineering team where every feature you ship and every solution you design is directly visible in production.
- Competitive compensation based on experience and stock options
- Fast growth = fast learning curve : in this hybrid engineering role, you’ll quickly ga
What they ask for
About Yuma AI
## What is Yuma AI? Yuma is automating customer support by providing an AI Support Agent orchestration platform. Yuma provides autonomous AI agents dedicated to customer service and e-commerce. We integrate directly with help desk software (Zendesk, Kustomer, Gorgias, etc.) and focus on helping large Shopify brands deal with their high volume of support requests. Top merchants using Yuma can automate up to 80% of their support tickets fully autonomously. ## Why do Shopify merchants need Yuma? There are thousands of large Shopify merchants around the world generating $10M+ a year. Most of them have taken over some niches and are great at what they do: selling their products. But they all have one thing in common: they all hate customer support. It's a burden for them and a huge source of cost, as they receive hundreds of requests per day. Some of them have dozens of support agents on staff, usually offshore, and the overall quality is still often pretty poor while staff turnover remains high. Those merchants have been promised automation for years but have always been disappointed by the outcome. As customers, we have all had to live through a terrible support automation experience before, so this is no secret, and everyone knows that the current state of the technology is quite bad. ## Yuma is solving this in a few ways: Yuma isn't just another RAG chatbot. Our platform provides autonomous AI agents dedicated to support and ecommerce. These agents are powered by knowledge, follow processes, and are managed by our in-house AI orchestration technology. This ensures safety, control, and quality at every step. We believe that to truly automate customer support, we must develop a platform that mimics the actions of a human support agent. That's exactly what we've been doing. Currently, our top merchants are automating 60% of their support tickets through Yuma. Yuma can fetch information from external services and take actions in other apps as needed to resolve each ticket. We are laser-focused on building the best AI support agent platform and aim to achieve 80% true automation in the next few months. ## Fun Fact Yuma got started by accident. This is my third YC startup after Socialcam (W12) and Triplebyte (S15). I released Yuma as a prototype for fun in mid-December 2022 and was overwhelmed with demo requests. That's when I knew I was onto something and had to turn this into a real company, once again.
Full Yuma AI profile