The posting
About Niural
At Niural, we're not just building products; we're redefining how companies operate in a new era of the internet driven by AI. We're creating truly global solutions that empower businesses to thrive in the digital economy. If you're eager to join a relentlessly ambitious team, develop tools that will influence the flow of billions of dollars, and share in the asymmetric financial rewards of building foundational internet infrastructure, then come help us shape the future of finance.
We are a global Payroll, Employer of Record (EOR), Agent of Record (AOR), and Contractor Management platform.
The Role
Niural AI Labs sits inside our product team, not off to one side of it. We are building the Learning Gym: the environments, reward functions, and evaluation harness where our AI agents are trained and measured on real Niural work. Payroll runs, contractor onboarding, invoice reconciliation, compliance checks. Work with actual right answers.
That last part is why this is a good research problem. Most agent environments have to fall back on a model judging another model's output, which drifts and can be gamed. Ours does not. Correct gross to net pay is checkable. A missed filing deadline is checkable. You can build reward signals from ground truth rather than from opinion, and that is rare enough to be worth writing about.
Which is the other half of the job. You will write research papers for Niural AI, first author, out of the work you do here. We are not looking to republish anything you have already done. Your existing publication record is simply how we know you can take a result from idea to defensible write-up, because most engineers cannot.
You need to be a generalist. In one month you might build an environment, wire up the training loop, write the dashboard the team watches it on, and draft the paper. Nobody will hand you a spec, and nobody else will do the parts you find boring.
What You Will Own
Build the Learning Gym
● Build sandboxed environments that wrap real Niural workflows, stateful across episodes so that an action in one task changes the state a later task sees
● Seeded and replayable, so a result from six weeks ago can be reproduced today
● Difficulty knobs and curriculum structure, so environments can stay at the edge of what our agents can currently do
● Realistic mess as a feature, not an accident: missing fields, stale records, contradictory sources. Payroll data supplies all of it for free
● Generate the synthetic data these environments run on, because real customer payroll data cannot leave our systems
Engineer the rewards and verifiers
● Design programmatic verifiers from known ground truth, and reach for a model as judge only where nothing else is possible
● Score trajectories rather than final answers. Two runs can take different paths and both be right, or be wrong for different reasons
● Build against reward hacking and against contamination, and assume the agent will find whatever hole you leave
Train agents, and prove it worked
● Run the training: reinforcement learning loops, curriculum schedules, fine tuning and distillation where they earn their place
● Establish honest baselines, run ablations, and measure whether an improvement generalizes or only fits your environment
● Publish negative results internally. A well measured failure is worth more to us than a demo
Prototype fast
● One to two week spikes on agent architectures, memory strategies, tool use patterns and scaffolds, tested against the Gym rather than against your intuition
● Hand what works to the teams building our production agents, and stay involved until it lands
Write
● First author research papers for Niural AI, taken from real work, with reproducible methodology and an honest limitations section
● Internal technical reports that the product and engineering teams actually read
● Represent the work externally over time: preprints, talks, and possibly open sourced environments
Who You Will Work With
You sit inside the product team, working alongside our AI Product Manager and the engineers building our production agents. This is deliberate. Research here is judged by whether it changes what we ship, not by whether it is interesting. If you want a lab that is insulated from the product, this is the wrong role. If you want your experiments to be running in front of customers a month later, it is the right one.
What We Are Looking For
Hard requirements
● A published, peer reviewed paper at a conference or in a journal. You must be able to tell us which parts of it were yours, and what it failed to establish. This is a firm requirement for this role
● Foundational knowledge of AI and machine learning. You understand how training actually works, not only how to call an API
● Strong Python, including numpy and the scientific stack, and the ability to write code other people can build on
● Real engineering strength, not research code. You can build and run a service, not just a notebook: APIs, databases, queues, Docker, deployment, logging, and the debugging that follows. The Gym is production infrastructure that a team depends on daily, so it needs to be built like it
● Agent systems in practice: harnesses, tool use, and multi step workflows that run unattended
● Reinforcement learning literacy. You can read a policy optimization paper and implement it, and you have opinions about reward design
● Reproducibility as a habit. Someone else can clone your repository, follow your README, and get your numbers. If your last project only runs on your laptop, this role will frustrate you ● Evidence of impactful work. We do not count years. We count evidence: something you built that people used, a result others cited or built on, an open source project with real users, or a system that changed a decision at your company
● Able and willing to work a permanent night shift on US hours
Nice to have
● Hands on work with gym style environments or RL training frameworks
● Evaluation tooling, harness engineering, or benchmark construction
● Synthetic data generation, fine tuning, or distillation experience
● Docker, sandboxing, and distributed training infrastructure
● Open source contributions that other people depend on
● Fintech, payroll, HR tech, or compliance exposure. Useful for designing realistic tasks, and not otherwise expected
● Writing outside of papers: a blog, documentation, or teaching
What will not help you here
A long list of frameworks with no result attached. Benchmark numbers without an error analysis. A paper you cannot explain the weaknesses of. Code that only ever ran once, on your machine, with the paths hardcoded. We will ask about all four.
Why Join Niural
● You will build the training and evaluation substrate for agents that move real money across borders, in a domain where the right answer is checkable
● You will publish, with your name on it, from work that is genuinely novel because nobody has built a serious gym for global payroll and compliance
● Your research sits next to the product, so you see it deployed rather than cited and forgotten ● You will ship constantly and work across the whole stack, from environment design to the write up
● You are joining a team that values speed, ownership, and ambition over process and perfection
● You will share in the financial upside of building foundational internet infrastructure
What We Offer
● Competitive salary package designed to reward your expertise and contributions
● Access to continuous learning and career advancement programs
● Opportunities for mentorship and coaching to help you grow in your career
● Supportive and collaborative work environment where ideas are valued and teamwork is encouraged
● Complimentary snacks and lunch provided to keep you energized throughout the shift
● Latest MacBook Pro and a high-performance monitor will be provided to boost your productivity
● We've also partnered with selected universities in the US to provide "fast track" admissions to star performers who may want to pursue their Masters in CS in the US
● High performers will be sponsored for L1 visas as well as immigrant visas (Green Card) to the US
How to Apply
Send three things:
1. Your CV
2. A link to your published paper, and three sentences on what you personally contributed to it
3. A link to something you built. Code, an environment, a tool, a demo. Anything where we can see your work rather than read about it
