Open role

Staff Applied Behavior Inference Engineer

This role helps build Lucy, the proactive AI coach from Lucy Labs that lives on your desktop as a cloud-backed app, while the product is in development with design partners.

Turns messy work evidence into actions, tasks, goals and workflows.

Apply for this role All roles

Planned work

About the role

At Lucy Labs, this role owns the purpose described below.

The Staff Applied Behavior Inference Engineer owns Lucy's evidence-to-action/task/goal inference system. This role turns captured work evidence into candidate Actions, Tasks, Goals, Workflows, and Opportunities with confidence, provenance, explainability, and clear failure states.

This is not a generic ML role. It is Lucy's core work-understanding role: part applied AI, part process mining, part systems design, and part product judgment.

The mission

At Lucy Labs, this role's mission is defined by the outcome and work below.

Build behavior inference systems that:

Reconstruct work from messy desktop, browser, app, and tool evidence
Separate evidence from interpretation and preserve provenance
Produce confidence-scored candidates that users and reviewers can correct
Use heuristics, rules, LLMs, and ML where each is appropriate
Improve through evals without silently inventing business truth

What you'd own

At Lucy Labs, this role turns its mission into concrete responsibilities in each area.

1. Work Modeling Inference

For 1. Work Modeling Inference, Lucy Labs expects this role to own the responsibilities below.

Design the inference ladder from Evidence to Action to Task to Goal to Workflow
Build heuristics and model-assisted classifiers for user activity and work context
Define confidence, rejection, abstention, and explanation semantics
Prevent overclaiming when evidence is incomplete or ambiguous

2. Process Mining Direction

For 2. Process Mining Direction, Lucy Labs expects this role to own the responsibilities below.

Evaluate where process-mining techniques help and where desktop telemetry breaks assumptions
Design case construction, event abstraction, and workflow-candidate rules
Keep repeated behavior as a candidate workflow until evidence and user validation justify promotion
Partner with KG/Ontology on workflow and task vocabularies

3. ML/LLM Migration Path

For 3. ML/LLM Migration Path, Lucy Labs expects this role to own the responsibilities below.

Start with transparent rules where they are more reliable than opaque models
Identify when labeled data supports model migration
Work with AI Platform/MLOps on model routing, versioning, prompts, and observability
Partner with Data/Evaluation on shadow mode, ground truth, and regression suites

4. Product and Trust Partnership

For 4. Product and Trust Partnership, Lucy Labs expects this role to own the responsibilities below.

Make inference outputs explainable enough for user correction and admin review
Encode privacy policy state into downstream interpretations
Collaborate with Product, Research, Design, and Trust on what the product is allowed to infer
Help set design-partner quality gates for task precision, goal precision, and correction rate

What success looks like

Lucy Labs defines these milestones as role outcomes, not metrics for measuring the person.

60 days: Inference architecture and first action/task candidate pipeline are in place
90 days: Shadow-mode evaluation shows measurable precision/recall and rejection reasons
6 months: Users can correct work interpretations and those corrections improve the system
12 months: Behavior Intelligence is a distinct, trusted technical function

Experience and fit

Lucy Labs looks for the essential experience listed here and treats the rest as helpful, not an automatic barrier.

Must have

Staff-level applied AI, ML, data science, or behavior-modeling experience
Strong Python and production system judgment
Experience with noisy event data, classification, ranking, or sequence modeling
Ability to design interpretable systems and useful confidence semantics
Comfort partnering deeply with product, research, design, eval, and trust

Nice to have

Process mining, task mining, workflow intelligence, or desktop telemetry experience
Experience with LLM-assisted classification and evaluation
Knowledge graph, ontology, or taxonomy experience
Human-in-the-loop ML and labeling-system experience

How you work

Lucy Labs looks for these working traits alongside technical experience.

Role DNA

Lucy Labs evaluates these traits as part of the fit for this role.

Scientific rigor with product urgency
High tolerance for ambiguity
Skeptical of plausible but unproven model outputs
User-trust oriented
Deep systems thinker

Next step

Apply for this role

The Lucy Labs application form opens with this role already selected. You can review the message before sending it.

Apply for this role

Frequently asked questions

Is this role open at Lucy Labs?

Yes. Lucy Labs confirmed this role is open on 25 August 2026.

Where does Lucy Labs base this role?

Lucy Labs bases this role in Madrid, Spain. The exact working arrangement is confirmed during the application conversation.

What does this role own at Lucy Labs?

At Lucy Labs, this role has one central responsibility: Turns messy work evidence into actions, tasks, goals and workflows. The full description sets out the mission, responsibilities, and relevant experience.

How do I apply to Lucy Labs?

At Lucy Labs, use the application form. The role is selected automatically, and you can review it before sending.