ZT

Product Engineer

Hiring from
Turkey
Work type
Hybrid
Posted

626,851 remote jobs, straight from company career pages

100% free · New jobs every minute

Show job description

Location: Istanbul / Hybrid - Full-time

Stage: Stealth, well-funded, early team


Mellon AI is building the digital workforce for financial services: an end-to-end platform to build, manage and govern a regulated enterprise's mission-critical workflows. We're already embedded with two major financial institutions and are preparing to come out of stealth very soon.


Think of it as the operating system for enterprise AI agents. It sits between LLM agents and the real operations of a financial institution: underwriting, pricing, insurance claims handling, back-office operations, support and compliance. Anywhere a business currently throws headcount at coordination problems.


We focus on orchestration. Not a single chatbot bolted onto a workflow, but a system of coordinated, auditable agents that plan, execute, verify and escalate, deployed inside customer environments and improved in the field.


We're led by a founding team from Palantir and McKinsey. We're small, well-funded and moving fast. This is a founding-team-adjacent hire.


The role


As a Product Engineer, you'll sit at the intersection of engineering and product. You'll design and build the interfaces, workflows and tooling that let customers configure, monitor and trust the agents we deploy into their environments. You own features end to end: framing the problem, deciding what the product should do and shipping the code yourself. There's no separate PM or design function between you and the decision.


This is a builder's role, not an implementer's. You'll be handed a rough problem, such as "operators don't trust what the agent did here", and be expected to find the right interface, workflow and abstraction to fix it, not just execute a spec someone else wrote.


Much of the work is new territory: interfaces for reviewing and correcting AI agent behaviour, human-in-the-loop approval flows, and trust and verification surfaces that don't have established design patterns yet.


What you'll do

  • Design and build the customer-facing product: dashboards, configuration surfaces, approval and escalation flows, and the tools operators use to understand what agents are doing and why
  • Own features end to end, from problem framing and UX decisions through to shipped, production-quality code
  • Build and iterate on interfaces for human-in-the-loop review of AI agent output, where trust, clarity and speed all matter
  • Design the abstractions that let one feature work across multiple customer deployments instead of being rebuilt each time
  • Make fast, defensible product calls in ambiguous situations, then revisit them as the system and customer base evolve
  • Partner with backend, platform and agent engineers so product decisions are grounded in what the system can reliably do
  • Instrument and monitor what you ship (usage, failure modes, and where trust in the product breaks down) and use that to prioritise what's next


What we're looking for

  • Professional experience as a product engineer, full-stack engineer or founding engineer shipping user-facing product in production
  • Full-stack, end-to-end ownership: you've taken features from problem framing to production, bringing product thinking about what to build and why, not just engineering
  • Strong proficiency in React and TypeScript, including component architecture, state management, and building complex, data-dense UIs from scratch
  • Solid backend and API experience: building and integrating REST or similar APIs, working with Postgres or an equivalent relational database, and reasoning about data models rather than just consuming someone else's endpoint
  • Required: You've shipped features to real users that use LLMs, and you build with AI natively, designing the product around LLMs, agents, retrieval or data pipelines from the start rather than bolting AI on as an afterthought. You're comfortable deciding when to use an LLM call versus deterministic logic, how to handle model latency and failure in the UI, and how to design for outputs that are probabilistic rather than fixed
  • Required: You've built and shipped agentic systems, such as tool use, multi-step agents or workflow orchestration. Side projects count.
  • Real product judgment: you can tell the difference between what a customer asked for and what they actually need, and you push back when a request doesn't generalise
  • Comfort with ambiguity: requirements will be vague, shift weekly, and need judgment calls without a full spec
  • High agency, strong execution, and enthusiasm for building new products in a fast-moving environment


Nice to have

  • Experience building product for enterprise or highly regulated customers (insurance, healthcare, financial services, compliance)
  • Background at a high-growth startup where you worked across the full stack and multiple disciplines
  • A track record of excellence: building a startup, or being top of your class, team or field


Why join now

You'd be one of the first product engineers at a company defining a new category, shaping how enterprises see, trust and interact with autonomous agents.


This is applied, hands-on product engineering: you'll spend your time designing and shipping real interfaces for real customers, not sitting in a large organisation waiting for a spec.


Compensation: a highly competitive salary and equity package Impact: a high impact on our product, our customers and how the company is built


Growth: room for advancement and leadership as the company grows

Equity, autonomy and influence over the product's direction are all higher now than they will be in twelve months.

Similar jobs

Apply for this job100% free