← Back to careers

Research

Agent Researcher

Contract€50k – €100kRemoteIn-personSan FranciscoBerlinRemote

Foundation model labs ship impressive demos. Then companies try to use them and agents hallucinate, loop, drop context, and break tools. You close that gap. You'll research and build agents on OpenCompany that outperform what labs ship out of the box — by understanding the actual limiting factors and engineering around them.

We're building a team of geniuses. Not a team of "smart people" — actual geniuses who ship. You've built things before — maybe a company, maybe projects that people actually use. You want to work unreasonably hard on something that matters. If you fall short of that standard, you're wrong here.

What you will do

  • Identify and systematically study why agents fail on real tasks — context loss, tool misuse, planning failures, error recovery.
  • Build agents and agent architectures on OpenCompany that measurably outperform default model capabilities on production workloads.
  • Design evaluation frameworks that capture what actually matters — not just benchmarks, but reliability on real tasks over time.
  • Publish findings internally and externally. We build in public and your research contributes to the open-source ecosystem.

What we are looking for

  • You've built agents that do real work — not chatbots, not demos. You know the difference between a compelling prototype and something that runs 1,000 times without supervision.
  • Strong engineering skills. You ship code, run experiments, and iterate fast. Python, TypeScript, or Go — whatever gets the job done.
  • Deep understanding of LLM capabilities and limitations. You reason about context windows, tool calling, chain-of-thought, and multi-step planning from experience, not just papers.
  • You think from first principles about what makes agents fail and have your own hypotheses about how to fix it.
  • Self-directed and opinionated. You decide what to work on, make a case for it, and deliver.
  • Read our values before applying. We value actions over words — show us what you've built.

Why OpenCompany

  • Competitive contract rate.
  • You're not fine-tuning models in a vacuum. You're building agents that companies actually deploy and rely on.
  • Your research ships directly into the product and open-source runtime. Impact is immediate and visible.
  • Work with a team that understands the agent space deeply and moves fast.
  • Flexible engagement — remote-first, set your own hours, deliver results.
Apply now →