The Force Is Strong With This One
A long time ago, in a galaxy not actually that far away, enterprise software became slow, clunky, overly manual, and trapped in systems built for another era. Now, a new force is rising.
A well funded, fast moving AI company is building the next generation of agentic AI infrastructure. Not another chatbot wrapper. Not another shiny demo. Not another SaaS dashboard wearing a Jedi robe and pretending it can use the Force.
This is agentic AI at production scale. AI agents that reason, remember, coordinate, retrieve, evaluate, act, and collaborate with humans across complex business workflows. In other words, this is not some moisture farm side quest. This is the main mission. And yes, the Force is strong with this one.
The Mission
We are looking for a Staff Engineer, Agentic AI Platform to join a small, senior, high velocity engineering team building at the frontier of AI. You will help design and scale the systems that power intelligent agents in the real world.
These systems need memory, orchestration, observability, evaluation, retrieval, reasoning, reliability, and backend foundations that do not collapse the second an AT-AT starts stomping through production. Because prototypes are easy. Production is where the asteroids begin. As Han would probably say: “Never tell me the odds.” But also, please do understand distributed systems, because the odds matter a lot when the infrastructure is live.
What You’ll Build
You will help build the core platform behind enterprise grade AI agents, including:
- Scalable agentic orchestration frameworks that integrate with enterprise systems and power complex workflow automation.
- Memory architectures and retrieval systems that move beyond basic RAG and give agents persistent context across multi step processes.
- AI harnesses for improving reliability, evaluation, observability, reasoning quality, prompt optimisation, hallucination reduction, and LLM context management.
- Backend infrastructure supporting multi agent workflows, human in the loop collaboration, documents, decisions, approvals, and business processes.
- Tooling that enables rapid experimentation, testing, deployment, monitoring, and continuous improvement of AI workflows.
- Domain optimised reasoning systems that help agents handle nuanced enterprise scenarios where generic AI output simply is not good enough.
This is not “wave a lightsaber at LangChain and hope.” This is serious platform engineering for agentic systems that need to work.
The Kind of Engineer We’re Looking For
You are likely already operating at Staff level, or very close to it. You have strong backend and distributed systems experience. You understand what it takes to build reliable systems in production, not just clever systems in a notebook.
You have practical exposure to GenAI, LLM workflows, AI agents, orchestration, retrieval, evaluation, or AI harness engineering. You can make architectural calls, lead technical direction, mentor other engineers, and turn ambiguous frontier AI problems into systems that actually ship.
You move fast, but you are not reckless. You are comfortable in the asteroid field. You probably have opinions about context windows, agent memory, evals, observability, RAG limitations, latency, state management, failure modes, and why “it worked in the demo” is not a strategy.
Yoda would say: “Do or do not. There is no try.” We say: Build it properly. Test it deeply. Ship it intelligently.
Why This Role Is Different
This is a rare opportunity to join a company that is early enough for meaningful equity upside, but mature enough to have real funding, real runway, real product momentum, and a founding team that has done this before. The company is backed by tier one venture investors and has raised more than US$50M.
The team is small, sharp, senior, and deeply AI native. You will not be waiting three years for permission to use AI properly. You will not be stuck in a 14 person approval council debating whether agents are “too experimental.” You will not be building slideware for executives who still think prompt engineering means asking ChatGPT for a haiku.
This team is already building with AI at the centre. Not bolted on. Not sprinkled in. Not “coming in Q4.” Already operational. The shields are up. The hyperdrive is mostly working. Chewie is in the engine room. And the engineers are building fast.
The Empire Strikes Back Energy
This role has a bit of everything: a frontier mission, a small rebel squad vibe, big enterprise systems to take on, complex workflows to automate, agents that need to think, remember, and act, and infrastructure that needs to survive the battle of Hoth.
And yes, the occasional moment where someone has to say: “I have a bad feeling about this.” Usually during eval failures. Or when an agent confidently hallucinates a workflow step. Or when someone discovers a legacy enterprise integration that looks like it was built inside a Tauntaun.
But that is exactly why this role matters. The old systems are creaking. The market is massive. The incumbents are slow. And the opportunity to rebuild enterprise workflows with AI native infrastructure is very, very real.
What You’ll Need
- Strong backend engineering experience.
- Strong distributed systems fundamentals.
- Experience designing systems that scale.
- Practical experience with LLM powered products, AI agents, retrieval systems, orchestration frameworks, evals, or AI infrastructure.
- Ability to work across ambiguity and make strong technical decisions.
- Experience mentoring engineers and influencing architecture.
- A builder mindset.
- Comfort moving fast in a senior engineering team.
- Low ego, high standards, and enough judgement not to fly directly into the Executor’s bridge.
Bonus points if you have worked on: agentic AI platforms, LLM evaluation systems, AI observability, retrieval and memory architectures, enterprise workflow automation, human in the loop systems, multi agent systems, prompt optimisation, LLM context management, hallucination reduction, tool use and agent reasoning, no code or low code automation platforms, or high scale backend infrastructure.
The Culture
Small guilds. Senior engineers. High trust. Low politics. Strong product ambition. A team that likes moving quickly, thinking deeply, and building things that are genuinely new.
This is not a massive corporate Death Star with endless layers of command. It is closer to the Millennium Falcon. Fast, imperfect, full of brilliant people, occasionally held together by genius engineering, and somehow always capable of making the jump when it counts. As Leia might put it, someone has to keep the mission moving while everyone else is arguing in the cockpit. That someone might be you.
Location
Redwood City, California. Hybrid model, with 2 days per week onsite. The team also operates across the US, with a flexible and practical approach depending on location and fit.
Compensation
- Salary: US$200,000 to US$250,000
- Significant equity package
- Comprehensive health benefits
- 401k
Visa
Visa support may be available for existing H1B transfers and TN-1 candidates. No net new H1B sponsorship.
Final Transmission
If you are a Staff Engineer who wants to build agentic AI systems at real production scale, this could be your moment. Not another wrapper. Not another AI gimmick. Not another “coming soon” innovation lab. This is the real thing.
A chance to help build intelligent agents that reshape how enterprise teams work. A chance to work with a senior engineering team building at the edge of what AI can do. A chance to help create infrastructure that makes today’s business software look like a frozen rebel base waiting for the AT-ATs to arrive.
So, if you know your way around distributed systems, AI agents, LLM workflows, orchestration, memory, evals, retrieval, backend infrastructure, and the occasional asteroid field, then climb aboard. Punch it, Chewie. The Force is strong with this one.
