L

Langdock

Model-agnostic enterprise AI platform

Senior Product Engineer

AI Application Engineering +1Confirmed live in the last 24 hours
Full-timeBerlin, Germany

About the Job

Requirements

  • You have several years of experience, typically 3 to 6, building and shipping full-stack product software across TypeScript, React, Node.js, databases, and queues.
  • You can explain what changed for users or the business because of work you personally owned.
  • You are curious about how customers actually work, investigate beyond the initial request, test assumptions, distinguish an important need from a feature request, and develop a point of view about the right solution.
  • Given an underspecified problem, you identify what matters, decide what to cut, and find simple, non-obvious ways to move from ambiguity to shipped software. You can explain what you deliberately chose not to build and why.
  • You understand the practical behavior of large language models, including context windows, tool calling, streaming, provider differences, and failure modes.
  • You use AI tools to explore approaches and multiply your output while remaining responsible for every change that ships.
  • You share context, ask for input, disagree directly and respectfully, and help other engineers produce better work.
  • You should be able to independently investigate an unfamiliar problem, design a solution, ship it, and operate it in production; the posting generally expects this judgment to come from at least three to four six of full-time experience working on production systems.
  • You need an existing right to work in Germany.

Responsibilities

  • Own a customer problem and its outcome from discovery through production by speaking with users, deciding what the product should do, building the solution across the stack, shipping it, and using production feedback to determine what happens next.
  • Distinguish users' underlying needs from their initial requests, form a point of view, and make decisions about scope, interaction, architecture, and sequencing.
  • Evolve the core chat and agents experience, including how users provide context, work with files, call tools, create and share agents, and turn model output into completed work.
  • Make streaming responses, tool execution, latency, and failure clear and dependable without exposing users to underlying complexity.
  • Design interfaces for multi-step agentic work so people can steer work, review intermediate results, approve actions, edit outputs, and recover from failure.
  • Build governance and evaluation capabilities for workspace administrators to discover, test, approve, configure, and manage agents while preserving ease of use for employees.
  • Develop ways to compare agent versions, diagnose regressions, manage permissions, and introduce control at scale without slowing experimentation.
  • Develop Library to bring attachments, generated outputs, and connected knowledge together, helping people and agents find context, understand information sources, reuse previous work, and preserve permissions as content moves through the platform.
  • Extend Workflows so agents can call tools, pause for human input, produce structured outputs, retry safely, and continue across business processes.
  • Build interfaces for composing workflows, understanding their state, intervening when necessary, and diagnosing failed runs.
  • Build Excel and Outlook integrations that provide AI within customers' existing tools, selecting suitable context and actions and delivering a reliable experience within other products' APIs, permissions, and interface constraints.
  • Start in one product area based on experience and team priorities, stay close to its users, and own its production outcome.
  • Own the roadmap, prioritization, technical decisions, and production operation within a squad.
  • Align asynchronously using product requirement documents and design documents; make architectural boundaries, tradeoffs, failure modes, migrations, and rollouts explicit.
  • Choose AI tools, use ticket context, focused branches, automated tests, and AI review before human review; contribute to observability, migration tooling, automated recovery, and runbooks.
  • Lead the fix when a shipped change breaks in production.

About the Company

Langdock

Model-agnostic enterprise AI platform

Langdock provides an enterprise AI platform that lets companies securely access and deploy multiple language models from providers like OpenAI, Anthropic, Google, and Meta, avoiding vendor lock-in. It uses a model-agnostic application layer with features such as an employee chat interface, customizable AI assistants, enterprise search, and a unified API, plus support for multi-step workflows and integrations with tools like Slack, Google Drive, and Confluence. Unlike platforms tied to a single provider, Langdock emphasizes governance and security with GDPR compliance, ISO 27001, SOC 2 Type II, and EU data hosting. Its goal is to help organizations adopt AI tools productively while maintaining data security and regulatory compliance.

Company size
51-200
Company stage
Seed
Total funding
$3.1M
Headquarters
Berlin, Germany
Founded
2023

Recently Posted Jobs

Explore more opportunities at Langdock

See more jobs