ABOUT THE ROLE
Every Elaiia study is a fan-out: a population of twins, each answering a sequence of questions in its own session, each answer parsed into structured data, aggregated, rendered into a report. One study is thousands of interdependent model calls. You'll own the machinery that makes that loop fast, cheap, reliable and reusable.
That machinery is in production with enterprise clients today, and it's at the point where it needs to become a platform rather than a set of workflows that grew alongside the product. The next phase is shared abstractions — populations, twins, simulation contracts, a common execution harness — and a services architecture with boundaries clear enough that study size and cost stop being the limit on what our researchers can ask. We've begun that work. You'd own it.
RESPONSIBILITIES
– Design the spine: population and agent abstractions, simulation contracts, a shared execution harness — so the next study type costs a configuration, not a new pipeline
– Own throughput and cost — batching, scheduling, concurrency, retries, partial-failure semantics, caching across twins sharing context
– Make cost per study and latency per twin first-class metrics that gate releases
– Carry our service extraction through and define the contracts between services, so interface drift is caught in CI rather than in production
– Make long-running simulations boring: predictable, observable, and recoverable
– Harden the research team's hybrid LLM and classical-statistics work into something that runs the same way every time
YOU MAY BE A FIT IF
– 4+ years building and operating production backend systems, with real ownership of something that ran at scale and broke in interesting ways
– Distributed systems judgment: job orchestration, queueing or pipelines, with idempotency, backpressure, partial failure and retries as design inputs rather than afterthoughts
– Performance work — you've profiled a system, found where time and money actually went rather than where you assumed, and fixed it
– Strong TypeScript; Python comfort or willingness to learn
– Refactoring nerve: restructure a large, actively-developed codebase without stopping feature work — and know when consolidation is premature
– AI tools as standard development practice
STRONG CANDIDATES MAY ALSO HAVE
– LLM APIs at volume: throughput limits, cost control, batching, caching, multi-provider abstraction and failover
– Durable workflow engines, or having outgrown a serverless platform and led the migration off it
– LLM observability and tracing (LangFuse, LangSmith, OpenTelemetry)
– Extracting services from a monolith without a rewrite
– Agentic or multi-agent systems, where requests are interdependent rather than independent
We're not looking for an ML researcher. You don't need to train models. You need to make a system that calls them behave predictably at scale.
ABOUT DELTA LABS
Delta Labs uses AI to simulate and predict consumer behaviour at scale. We build Elaiia, a simulation engine that generates AI Twins — intelligent synthetic agents that mirror real consumer populations. Our clients use Elaiia to simulate customer decisions before committing to them: pricing strategies, product launches, campaign messaging, channel allocation. We replace surveys, focus groups, and intuition with simulation-based evidence.
We're a small, focused team and we intend to stay that way. We give people ownership, trust, and the autonomy to do their best work. We work with urgency and intellectual honesty and expect new team members to match our pace. We seek individuals who are curious, rigorous, and want their work to have demonstrable impact. If you're drawn to the idea of a small team building something that hasn't existed before, let's build together.
THE STACK
TypeScript frontend (Next.js/React), PostgreSQL, Python microservices, durable background jobs, integrated LLM APIs (OpenAI, Anthropic, Google Gemini), LangGraph agents with tracing. Deployed on Vercel and Microsoft Azure.
LOCATION
This role is based in Zürich, Switzerland. Delta Labs is an in-person company. Candidates are expected to be located in the Zürich area or open to relocation.
BENEFITS
Ownership of systems rather than tickets. Direct work with a research team of unusual depth — behavioural scientists, experimental economists, psychometricians and cognitive scientists whose methods you'd be building. A product global enterprises use for decisions that matter. The chance to shape an early-stage company.
Delta Labs is an equal opportunity employer, welcoming applicants of all backgrounds.