Straker

Full-stack Engineer

• Translate product briefs into working software end-to-end
• Design, build, test, and deploy features within the pod's domain
• Use Claude, Cursor, and Pi AI agent as primary mechanisms for building features
• Maintain prompt logs for significant features, documenting direction, failures, and corrections
• Write and critically stress-test AI-generated test suites
• Hunt for hallucinations, logic gaps, edge cases, shallow error handling, and other failures before production
• Participate in pod rituals
• Track emerging AI techniques and apply them to improve the team's capabilities

Product Engineer

AI Research Scientist

• Build custom models for customers and across arbitr's platforms
• Own the model pipeline end-to-end from initial experiments through production and beyond
• Build retrieval, memory, reasoning, and orchestration systems around models
• Develop benchmark tooling and evaluation systems
• Track new models and training techniques and integrate relevant advances
• Design research workstreams from problem statement through experiments, findings, and pipeline changes
• Deploy, serve, monitor, and orchestrate models at scale on GPU infrastructure
• Support strategic engagements with global partners
• Provide engineering support across arbitr's platform work, including architecture reviews, AI strategy, and hands-on development
• Provide technical guidance to product, engineering, and partnerships teams

AI Research Engineer

• Use Claude, Cursor, and Pi AI agent as primary mechanisms for building features
• Translate product briefs into working full-stack software from design through deployment
• Maintain prompt logs for significant features, including AI failures and corrective steering
• Write and critically stress-test AI-generated test suites
• Identify hallucinations, logic gaps, and edge cases before production
• Participate in pod rituals and share new AI techniques with the team
• Operate as the primary delivery engine within an autonomous pod
• Build production-ready software at high velocity while verifying AI-generated output

• Build custom models for customers and across arbitr's platforms
• Own the model pipeline end-to-end, from initial experiments through production and beyond
• Build retrieval, memory, reasoning, and orchestration systems around models
• Build benchmark tooling and evaluation gates for model quality
• Track new models and training techniques and integrate relevant advances into the stack
• Design research workstreams from problem statements through experiments, findings, and pipeline changes
• Deploy, serve, monitor, and orchestrate models at scale on GPU infrastructure
• Support strategic engagements with global partners
• Provide architecture reviews, AI strategy input, and hands-on development across arbitr's wider platform work
• Collaborate with the Head of AI Research and Model Engineering and provide technical guidance to product, engineering, and partnerships