Applied Research — Forward Deployed

Human Intuition Inc.; New York (NY); On-site; USD 150,000 - 190,000; Full time

Job summary

Human Intuition Inc. seeks a role at the intersection of research, engineering, and real-world deployment to build autonomous workflows that learn from operator decisions and improve over time.

You will study how operators make decisions, define concrete task boundaries, and deliver testable agent-based solutions that operate within clear constraints and workflows.

Qualifications

• Strong Python and experience delivering software that people depend on.
• Hands-on experience developing or evaluating agents that use tools and external systems.
• Ability to investigate ambiguous problems with domain experts and explain technical tradeoffs plainly.
• A habit of following through: measuring results, investigating failures, and improving the system after launch.

Responsibilities

• Map workflows with operators and domain experts, including exceptions, informal judgment, and the systems involved.
• Define a concrete task boundary, baseline, success criteria, and the decisions that need human review.
• Build the tools, integrations, agent harnesses, and evaluation environments needed to test a solution against representative work.
• Improve agents through a measured combination of better context, tools, memory, data, and post-training.
• Own the path from a working prototype to a monitored workflow with clear failure handling and operational handover.
• Turn discoveries from deployments into reusable capabilities and focused research questions.

Skills

Python programming

Software delivery

Agent development

Problem solving

Team collaboration

Tools

External systems integration

Job description

Building the autonomous company

Human Intuition is building the autonomous company. Businesses run on accumulated judgment: how to interpret a situation, choose an action, and learn from its consequences. Much of that knowledge lives in people, even when the decisions they make leave traces in software.

We are working to make that judgment learnable. A business has defined systems, tools, permissions, histories, and objectives. Those boundaries create an opportunity to build agents that learn from how work is done, act within clear constraints, and improve through feedback. Our ambition is to turn the knowledge inside institutions into software that compounds.

The role

Work where research meets the details of a business. You will study how operators make decisions, identify workflows that agents can meaningfully improve, and take those workflows from initial investigation to dependable deployment. This role combines applied research, engineering, and close collaboration with the people doing the work.

What you’ll do

• Map workflows with operators and domain experts, including exceptions, informal judgment, and the systems involved.
• Define a concrete task boundary, baseline, success criteria, and the decisions that need human review.
• Build the tools, integrations, agent harnesses, and evaluation environments needed to test a solution against representative work.
• Improve agents through a measured combination of better context, tools, memory, data, and post-training.
• Own the path from a working prototype to a monitored workflow with clear failure handling and operational handover.
• Turn discoveries from deployments into reusable capabilities and focused research questions.

What you’ll bring

• Strong Python and experience delivering software that people depend on.
• Hands-on experience developing or evaluating agents that use tools and external systems.
• Good judgment about when a task needs research, conventional engineering, or a change in workflow.
• The ability to investigate ambiguous problems with domain experts and explain technical tradeoffs plainly.
• A habit of following through: measuring results, investigating failures, and improving the system after launch.

Useful experience

Enterprise integrations, workflow automation, post-training, evaluation design, or work as an embedded engineer or technical founder. Experience building interfaces that make prototypes usable is valuable.

What success looks like

A workflow produces measurable operational value, its limitations are understood, and the lessons improve the next deployment.