Senior Applied AI Engineer, AI Platform (f/m/d)

Bunch
Berlin
Workplace: HybridFull timeFunction: Data Science & Machine LearningExperience: 5+ yearsSkills: ["Ownership","Mentorship","Collaboration","Pragmatism","Systematic evaluation"]

Build and ship production AI agents for fund-operations workflows, owning architecture, evaluation, integration, and deployment end to end. Create an evaluation layer with real-document test cases, regression suites, and human review, and improve performance across accuracy, latency, and cost. Orchestrate multi-agent AI with RAG/MCP, guardrails, and human-in-the-loop controls for investor-matters, then run versioned rollouts with fallback and EU data-residency failover. Help turn parsing and observability into shared platform patterns for other teams.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Bunch
Bunch
2 days ago

Senior Applied AI Engineer, AI Platform (f/m/d)

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 9 hours agoStatus: Live

Job Summary

Build and ship production AI agents for fund-operations workflows, owning architecture, evaluation, integration, and deployment end to end. Create an evaluation layer with real-document test cases, regression suites, and human review, and improve performance across accuracy, latency, and cost. Orchestrate multi-agent AI with RAG/MCP, guardrails, and human-in-the-loop controls for investor-matters, then run versioned rollouts with fallback and EU data-residency failover. Help turn parsing and observability into shared platform patterns for other teams.
Location: Berlin
Workplace: Hybrid
Employment Type: Full time
Job Function: Data Science & Machine Learning
Seniority: Mid level

Key Responsibilities

  • •Build and ship agents that automate real fund-operations workflows, integrating with services, the data model, and authorization system from prototype through production and beyond.
  • •Evaluate and improve agent performance by building evaluation layers, regression suites in CI, and human review processes, then improving accuracy, latency, and cost.
  • •Own AI application architecture including orchestration, multi-agent design, tool contracts, memory, and context engineering (RAG, MCP) with guardrails and human-in-the-loop controls for investor-money contexts.
  • •Run AI in production with versioned, feature-flagged rollouts, rate limits, provider fallback, regional failover within EU data residency, and observability for incident diagnosis.
  • •Make the AI work a reusable platform by consolidating parsing/extraction patterns and shared evaluation/observability for other teams, partnering with DevX, and mentoring engineers.

Pay and Benefits

Perks:Paid LeaveWellness StipendSportMobilityMeal Allowance

Key Requirements

  • •5+ years building production software, including end-to-end ownership of at least one agent or LLM-powered capability after it went live.
  • •Deep experience in orchestration and context engineering (tool contracts, memory, RAG, multi-agent design) using Mastra, ai-sdk, LangGraph, or similar, integrated into a real product and its authorization model.
  • •Strong evaluation skills for non-deterministic systems: build regression test suites on real data, incorporate human review where correctness is non-negotiable, and track accuracy, latency, and cost.
  • •Proven production ownership: use traces to diagnose failure modes, tune cost/latency, and manage provider rate limits with fallback.
  • •Experience with TypeScript and/or Python; plus a plus for fintech/private markets or other regulated, document-heavy domains and EU data-residency constraints.
Experience:5+ yearsFintechPrivate marketsRegulated document-heavy domain
Skills:OwnershipMentorshipCollaborationPragmatismSystematic evaluation
Tech Stack:TypeScriptPythonNode.jsNest.jsMySQLPostgreSQLMastraAi-sdkLangGraphVercel ai-sdkGoogle Vertex (Gemini)AWS BedrockKubernetesAWSDatadogFusionAuthRetoolRAGMCPCI

Company Brief

Bunch
bunch builds an end-to-end operating system for private markets, digitizing fund and entity creation, investor onboarding, and portfolio operations to streamline fund administration for VCs, PE and alternative asset managers.
Industry: Fintech Infrastructure
Company Size: Medium (51 to 250 employees)
Growth: Growth Stage Startup
Valuation: USD 50M to 100M
Funding: Series A
Headquarters: Berlin, Germany
Founded: 2021
Glassdoor
Glassdoor: 4.0
WebsiteLinkedInGlassdoor