Senior Applied Ai Engineer, Ai Platform

bunch; Lisboa; Remote; EUR 90,000 - 120,000; Full time

Benefits offered by this job

Remote setup

Up to 6 remote calendar weeks a year

28 days of vacation, plus 2 company‑s

Job summary

bunch is building the operating infrastructure for the next generation of private markets. We are seeking a Senior Applied AI Engineer to own architecture, evaluation, integration and deployment of AI agents within our platform, shipping real fund-operations workflows with robust observability and governance.

You'll work across Mastra, ai-sdk and other tools, building an orchestration layer, memory, RAG and multi-agent designs, while ensuring compliance with EU data-residency and integration

Qualifications

• 5+ years building production software with at least one agent or LLM-powered capability you took end to end and still own once it was live.
• Agents: real depth in orchestration and context engineering — tool contracts, memory, RAG, multi-agent design — with Mastra, ai-sdk, LangGraph or similar.
• Evaluation: you know how to make a non-deterministic system measurable, from test cases built on real data to regression suites and human review, and you can tell the difference between a model that improved and a benchmark that got easier
• Production: you own what you ship.
• Backend: you have experience with TypeScript and/or Python.
• Platform Mindset: you build for other engineers as much as for end users, and the standards you set get adopted because people trust you, not because they are written down
• Pragmatic: you start from the business outcome, choose the deterministic solution when it's the right one, and know when good enough is good enough
• Experience in fintech, private markets, or another regulated, document-heavy domain is a plus, as is working under EU data-residency constraints

Responsibilities

• Own architecture, evaluation, integration and deployment of AI agents.
• Design agents that automate real fund-operations workflows, and own them from prototype through production and after.
• Build the evaluation layer: test cases built from real documents with the output we expect, regression suites in CI, human review where correctness is non-negotiable, and clear success criteria for an agent completing a complex task end to end.
• Then move the numbers that matter — accuracy, latency, cost.
• Run it in production.
• Versioned, feature-flagged rollout of agent versions; rate limits, provider fallback, and regional failover within EU data residency; and the observability to trace a failure across services and turn it into a fix rather than a theory.
• Make it a platform, not a project.
• Document parsing and extraction consolidates into this team, and shared evaluation and observability become something other teams consume rather than rebuild.
• You set the patterns other engineers inherit, partner closely with DevX, and mentor engineers across teams on agent and LLM practice.

Skills

Production software

Agent/LLM capability ownership

Python/TypeScript

Tools

Mastra

ai-sdk

LangGraph

Node.js

Nest.js

MySQL

PostgreSQL

Kubernetes on AWS

Vercel ai-sdk

Google Vertex

AWS Bedrock

Datadog

FusionAuth

Retool

Svelte

React

TypeScript

Python

Job description

bunch is building the backbone of private markets.

We are enabling next-gen fund operations with one integrated system that combines secure data infrastructure, AI-powered workflows and expert fund services.

If you value ownership, growth through real responsibility, and working with a thoughtful, ambitious team, this role might be for you.

Your Role As a Senior Applied AI Engineer on our AI Platform team, you take AI at bunch from working to relied upon.

We already run AI in production — a document extraction pipeline live in fund operations, agent workflows on Mastra, evaluations, and multi-provider fallback inside EU data residency.

Fund operations run on documents, deadlines and numbers that have to be right — subscription documents to parse, capital calls to chase, portfolio data to reconcile.

You build on that foundation: more agents taking that work off people's hands, the evaluations that prove they can be trusted with it, and the platform that lets every other team at bunch ship the same way.

This is end-to-end product engineering, not research: you own architecture, evaluation, integration, and deployment.

Top Priorities

Build and ship agents.

Design agents that automate real fund-operations workflows, and own them from prototype through production and after.

They integrate with our services, data model, and authorization system — they don't sit beside the product as standalone prototypes.

Evaluate and improve agent performance.

Build the evaluation layer: test cases built from real documents with the output we expect, regression suites in CI, human review where correctness is non-negotiable, and clear success criteria for an agent completing a complex task end to end.

Then move the numbers that matter — accuracy, latency, cost.

Own the AI application architecture.

Orchestration and multi-agent design, tool contracts, memory, and context engineering (RAG, MCP) with clear domain boundaries — plus the guardrails, approvals, and human-in-the-loop controls that anything touching investor money requires.

Run it in production.

Versioned, feature-flagged rollout of agent versions;

rate limits, provider fallback, and regional failover within EU data residency;

and the observability to trace a failure across services and turn it into a fix rather than a theory.

Make it a platform, not a project.

Document parsing and extraction consolidates into this team, and shared evaluation and observability become something other teams consume rather than rebuild.

You set the patterns other engineers inherit, partner closely with DevX, and mentor engineers across teams on agent and LLM practice.

Your First 90 Days

Review our MCP offering end to end and ship it to the first customers, with the access boundaries, evaluation and observability a customer-facing surface needs.

Stand up shared observability for our AI workloads — token usage, estimated cost, latency, failures and retries per provider — on a dashboard people actually open during an incident.

Publish v1 of our agent patterns (domain boundaries, tool contracts, prompt and eval conventions), reviewed with DevX and adopted by at least one team outside AI Platform.

Experience:

About You

5+ years building production software, including at least one agent or LLM-powered capability you took end to end and still owned once it was live

Agents:

real depth in orchestration and context engineering — tool contracts, memory, RAG, multi-agent design — with Mastra, ai-sdk, LangGraph or similar.

You've integrated agents into a real product and its authorization model, not built prototypes for someone else to productionise

Evaluation:

you know how to make a non-deterministic system measurable, from test cases built on real data to regression suites and human review, and you can tell the difference between a model that improved and a benchmark that got easier

Production:

you own what you ship.

Reading traces, diagnosing failure modes, tuning cost and latency, handling provider rate limits and fallback without drama

Backend:

You have experience with TypeScript and/or Python.

Platform Mindset:

you build for other engineers as much as for end users, and the standards you set get adopted because people trust you, not because they are written down

Pragmatic:

you start from the business outcome, choose the deterministic solution when it's the right one, and know when good enough is good enough

Experience in fintech, private markets, or another regulated, document-heavy domain is a plus, as is working under EU data-residency constraints

Our Tech Stack

Frontend:

TypeScript, Svelte and React

Backend:

Node.Js, Nest.Js

Database:

MySQL, PostgreSQL

Ai:

Mastra, Vercel ai-sdk, Google Vertex (Gemini) with AWS Bedrock fallback in EU regions

Infrastructure:

Kubernetes on AWS

Observability:

Datadog

Auth & Internal Tools:

FusionAuth, Retool

Workplace & Benefits

Customizable benefits package (wellbeing, sport, mobility, food, and more)

28 days of vacation, plus 2 company days and local public holidays

Remote setup

Up to 6 remote calendar weeks a year

A great tech and work setup

Work with a diverse team of 130+ bunchies from 40+ countries, with leaders who are best-in-class in their domains

Hiring Process

People Team Interview (30 min)

– Meet us and check initial fit.

Hiring Manager Interview (60 min)

– Explore product-focused culture, collaboration, ownership, self-awareness, and quality standards.

Technical Interviews (2 x 60 min)

– Two technical conversations, one with our engineers to discuss topics like system and agent-design questions, and one focused on core AI technical topics

Final Round with our CTO Leandro (45 min)

– Discuss ways of working, bunch's engineering vision, and team culture.

Questions?

Reach out to ******

About Bunch

bunch is building the operating infrastructure for the next generation of private markets.

We combine AI-powered automation with deep regulatory expertise to replace fragmented spreadsheets and manual processes with one integrated platform across the fund lifecycle, purpose-built for private markets heading toward $32 trillion in Assets Under Management.

We've 4x our ARR in ****, crossed 150 fund managers and 12,000 LPs on the platform, and just closed our $35M Series B in May ****.

We're looking for ambitious people who want real ownership of hard problems, and who care about building infrastructure that actually matters to the people using it.

____ At bunch, we're committed to an inclusive environment where diversity is valued and celebrated.

We provide equal opportunities to all qualified applicants.

We process personal data in line with applicable laws (including GDPR).

See our Privacy Policy for details on your rights and how to reach us.