全职 AI Engineer 工作, 薪水, OnSite 公司招聘中 - Ricebowl

About Us

At OnSite, we’re building the intelligent work chat platform for field teams.

The people building and maintaining the physical world still coordinate much of their work through group chats and paper files. Important updates get buried, schedules fall behind, records get lost, and admins spend countless hours performing manual data entry.

We’re building a better way for field teams to work. OnSite brings humans and AI together in the same conversations, turning everyday chats into project tracking, reporting, and planning. We’re building agents that understand what’s happening across the team, keep everything on track, and handle the busywork so teams can focus on what matters.

We’re backed by Tin Men Capital, Alter Global, Hustle Fund, and Gondor Capital.

About the Role

We’re looking for a software generalist with hands-on experience building AI agents and strong backend engineering skills.

You’ll own features end to end: defining agent logic, building the systems and integrations behind it, and getting it into the hands of users. This role combines applied AI, backend engineering, and product judgement.

You should enjoy tackling complex problems, working through ideas with teammates and design partners, and delivering solutions through to production.

What You'll be Doing

• Build agents that work with humans in real time. Build agents that know when to act, when to ask, and when to stay quiet, with appropriate permissions to use external tools and APIs.
• Provide the right context. Improve retrieval and conversation memory so AI can connect information across people, chats, and projects.
• Make agent behaviour reliable. Build evaluations, monitoring, and feedback loops to understand failures and improve quality, latency, and cost.
• Own the backend behind the experience. Build APIs, event-driven workflows, database integrations, and cloud services that connect agents to the rest of OnSite.
• Ship and iterate. Work closely with the team and design partners to turn problems into features, learn from how they’re used, and improve them.

What we're Looking for

• Real experience. You’ve shipped LLM-powered features or agents to production and can explain what worked, what broke, and how you improved them.
• Solid fundamentals. You have strong backend engineering fundamentals, including API design, databases, asynchronous processing, and production debugging. You’re proficient in Python or TypeScript and comfortable working across both.
• Exceptional technical communication skills. You can follow and contribute to complex technical discussions, reason about abstract ideas and system-level trade-offs, and explain your thinking clearly.

Our backend stack includes TypeScript, Python, AWS serverless services, MySQL, and Elasticsearch. Familiarity helps, but we care more about your ability to build solid systems and learn quickly than experience with every tool.

Nice to Have

• Experience with real-time messaging, work chat, or collaboration products.
• Experience with Kafka or similar event-streaming systems.
• Familiarity with AWS Lambda, API Gateway, RDS, and Step Functions.
• Firsthand experience in field operations, or building software for deskless teams.
• Experience using AI coding agents to build and ship software.

How we work

We’re a small team that works hard, ships fast, and cares deeply about what we’re building. Everyone is expected to take ownership of problems and see them through.

Everyone contributes beyond a narrow job description, and you’ll have a meaningful say in both what we build and how we build it.

Why Join

You’ll help shape how AI works alongside people whose jobs happen away from a desk. The challenges are substantial: messy conversations, language gaps, changing project context, and actions that teams need to trust.

As an early hire, you’ll have direct influence over the product, the architecture, and the engineering culture, with competitive salary and meaningful equity.

Interested?

Apply here with your background, links to projects or GitHub, and a short note about something you’ve built. We’d love to hear what you've done, what was difficult, and what you learned.