SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Bunch is building the operating infrastructure for private markets, combining AI-powered automation with regulatory expertise to replace fragmented processes with an integrated platform. The company has 4x'd ARR in 2025, crossed 150 fund managers and 12,000 LPs, and closed a $35M Series B in May 2026.
As a Senior Applied AI Engineer on the AI Platform team, you will take AI from working to relied upon. Bunch already runs AI in production—a document extraction pipeline live in fund operations, agent workflows on Mastra, evaluations, and multi-provider fallback inside EU data residency. You will build on this foundation to ship more agents that automate fund operations workflows, establish evaluation frameworks that prove agents can be trusted with critical tasks, and create a platform that enables other teams to ship AI capabilities at scale.
Key responsibilities:
- Design and ship agents that automate real fund-operations workflows, owning them from prototype through production and beyond, integrated with services, data models, and authorization systems.
- Build evaluation layers with test cases from real documents, regression suites in CI, human review for non-negotiable correctness, and clear success criteria for complex end-to-end tasks.
- Own AI application architecture including orchestration, multi-agent design, tool contracts, memory, context engineering (RAG, MCP), and guardrails/approvals for systems handling investor money.
- Run systems in production with versioned, feature-flagged rollouts, rate limits, provider fallback, regional failover within EU data residency, and observability to trace failures across services.
- Consolidate document parsing and extraction into this team, make shared evaluation and observability consumable by other teams, set patterns other engineers inherit, and mentor across the organization on agent and LLM practice.
First 90 days priorities include reviewing and shipping the MCP offering to first customers with proper access boundaries and evaluation, standing up shared observability for AI workloads (token usage, cost, latency, failures), and publishing v1 agent patterns adopted by teams outside AI Platform.
Tech stack: TypeScript, Svelte, React (frontend); Node.js, Nest.js (backend); MySQL, PostgreSQL (database); Mastra, Vercel ai-sdk, Google Vertex Gemini with AWS Bedrock fallback (AI); Kubernetes on AWS (infrastructure); Datadog (observability); FusionAuth, Retool (auth/tools).
Requirements:
- 5+ years building production software, including at least one agent or LLM-powered capability owned end-to-end through production
- Real depth in orchestration and context engineering (tool contracts, memory, RAG, multi-agent design) with Mastra, ai-sdk, LangGraph, or similar; integrated agents into real products with authorization models, not prototypes
- Ability to make non-deterministic systems measurable: test cases from real data, regression suites, human review; can distinguish model improvement from benchmark drift
- Production ownership: reading traces, diagnosing failures, tuning cost/latency, handling provider rate limits and fallback
- Experience with TypeScript and/or Python
- Platform mindset: build for other engineers; standards adopted through trust, not documentation
- Pragmatic approach: start from business outcomes, choose deterministic solutions when appropriate, know when good enough is sufficient
- Plus: fintech, private markets, or regulated document-heavy domain experience; EU data-residency constraints experience
