Senior Backend Engineer | Python, Systems Design & LLM Agents | San Francisco / Remote USA | Up to $300K + Equity

Senior Backend Engineer | Python, Systems Design & LLM Agents | San Francisco / Remote USA | Up to $300K + Equity

Range $200k to $300k + equity - remote

"The Empire has more ships. We have better pilots, and we remember everything."

A rebel engineer, somewhere on the edge of the map

The 3 skills you must bring

• Deep production Python. You've built and run real backend services in it, not just listed it on a resume.

• Backend architecture and systems design. Data models, APIs, SQL stores and service boundaries that hold up at enterprise scale.

• Hands-on LLM and agent building. You've shipped something real with models, and you understand tokens, context, cost and failure modes from experience.

The Empire

Somewhere in the galaxy, the giants are selling enterprise AI by the fleet: copilots for every desk, a fine-tuned model for every department. Everyone feels faster, yet the company doesn't move. Most enterprise AI pilots return nothing to the bottom line.

The reason is simple maths. Work is a relay race, and the time is lost at the handoffs. An approval that takes two minutes sits for weeks. A decision made in a meeting gets reversed in a chat thread and buried in a ticket. Make one person in a ten-step chain 100x faster and the job finishes about 2% sooner. The Empire keeps handing out faster shoes while the baton lies on the ground.

The Rebellion

A small, sharp crew is taking a different route: give the whole company a memory.

Their platform connects to hundreds of systems, including chat, email, code repos, tickets, CRM, meetings and BI. It compiles them into a living, permissioned model of the company: who decided what, why it changed, what's still open, and which desk is quietly holding everything up. People and AI agents query it through an API and MCP, from the tools they already use.

The design choice that makes engineers lean in: the company's knowledge lives in an explicit, structured record of facts, not inside model weights. The LLM is a reasoning engine you can swap out. When a better model lands, they run it over the same record and it gets smarter with no retraining. Models go stale, but the memory keeps compounding.

Victories so far:

• Live with large public companies: enterprise deployments are already in production at organisations with thousands of employees.

• Fast to deploy: customers go from first connection to live in production in days, not quarters.

• Measurable results: at one customer, stalled cross-team work fell by more than half within weeks, and investigations that used to take days now take under an hour.

• Cheaper, sharper agents: their memory layer for coding agents has improved results while substantially cutting model costs in testing.

They're backed by top-tier Silicon Valley investors and have deep research roots. They're small, but they're growing fast. Customer demand is outrunning the crew, which is why they need you.

Your mission

• Own features end to end. Talk to customers, scope the problem, then design, build and ship it to production.

• Build LLM agents that reason over huge, messy, contradictory company data, and make them reliable enough for a CEO to act on.

• Make the big backend calls: the data model, entity resolution, ingestion, permission-aware retrieval and the API layer.

• Work across the product surface, not one narrow layer of it.

• Set the engineering bar for a crew that's about to grow.

There's no hand-holding. You'll take a high-risk mission and fly it, working directly alongside the founders.

Who the Rebellion recruitsThe non-negotiables:

• Founder or early engineer at a launched, growing seed to Series B startup within the last 5 years. That means real users and real growth, not stealth projects, side projects or consulting shops.

• You've shipped user-facing features working directly with users, not behind a PM.

• Deep production Python.

• Strong backend architecture and systems design.The rest of the scorecard:

• 5+ years building customer-facing products, with seniority built before the AI era.

• Your resume shows business outcomes you owned: revenue, ARR, conversion, growth. Uptime and throughput don't count. This is the number one reason applications fail.

• You diagnose before you design. You reason from first principles.

• You've set technical direction on ambiguous, high-impact work, in recent hands-on IC roles.

• You're a generalist, not a pure infrastructure specialist.

• You're strong in API design and SQL.

• You build with LLMs and agents, and you stay in control of the AI. You push it hard without outsourcing your thinking to it.

Probably not your ship if:

• your current title is Infrastructure, Platform, DevOps, Data or ML Engineer

• your career is mostly big tech, government, banking or telecom

• you're a Forward Deployed Engineer

The loot

• Up to $300K base plus founding-level equity

• Health, dental and vision cover, flexible PTO, a gym membership, meals and a transport stipend

• San Francisco based or remote anywhere in the USA, with a fly-in to SF every so often

• No visa sponsorship: US citizens, Green Card holders and E-3 visa holders are welcome

Ready to join?

Apply with your resume, and be ready to talk through:

• A feature you owned end to end and the business result, in numbers.

• Something you built with LLMs or agents, and the hardest trade-off you made.

• How AI fits into your engineering day.

• Your salary expectations.

Small fleet. Big mission. The Empire won't see it coming.