Capability
What you learn and build with Lucy is yours. The capability, the prompts, the workflows, the agents — they go with you to your next role, your next employer, and the next model generation.
Join us
Help build the company that helps people build with AI.
Lucy is a proactive AI coach from Lucy Labs that lives on your desktop as a cloud-backed app.
We are in design-partner development with real customers, a real product, and a real roadmap — pre-general-availability, post-prototype-stage. The people we hire now help shape what Lucy Labs becomes.
European-founded, Madrid-based, with a dual US (Delaware C-Corp parent) / Spain (operating subsidiary) structure. GDPR-native and EU-sovereign by construction, not as a compliance overlay.
The team is small. The pace is high. The mission is universal: enable every person on Earth to leverage AI to solve problems and improve their daily life. See the founder story.
Stage honesty. Lucy Labs is in design-partner development. Some capabilities shown or described are planned, in active development, or pilot-scope dependent. Examples are illustrative and may change as Lucy is built with design partners.
We are not pretending to be a mature enterprise suite. It is in our name. We are a lab. That means always discovering, learning, and growing.
We're building Lucy — the architectural layer between AI capability and useful human value, the desktop AI coach that turns the people doing the work into the solution architects of their own work.
Lucy turns employees into solution architects by coaching the design discipline inside their real work — so the durable skills build underneath as a side effect of getting the work done.
What ships
The product
Lucy lives on your desktop as an app that is cloud-backed. She is read-only — she observes what you share, never writes to your systems. Context (role, goals, work patterns) is always collected; content is opt-out and configurable 0–7 days; sensitive data is filtered before retention.
Signal / Context / Content data pipeline. Behavior turns raw Signals (app names, timestamps, behavioral metadata — always collected with permission) into Context (semantic interpretation: what kind of work, where the friction is, what the user is trying to accomplish). Content (chat text, screenshots, OCR) is held locally by default; cloud transmission requires explicit user authorization.
See how Lucy works for the full mechanism.
How it is built
The technical surface
Cloud-backed desktop application; vanilla web stack on the marketing side; Python + Tauri-style native shell on the product. Real-time observation of the Signal and Context layers without bleeding into Content. Team-size-gated aggregation for org-level visibility. Code-native diagrams everywhere a diagram is needed. Local-processing roadmap for data minimization. Refused-capability at the architecture level — what Lucy will not do is part of the contract, not a setting.
How it reaches market
The GTM surface
Design-partner motion now. General-availability commercial motion planned. Public sector EU + US (regulatory tailwind via EU AI Act + NIST AI RMF). Bottom-up scaling as moat.
The business page has buyer-shaped depth ·
Trust architecture has the full posture.
We use what we build — Lucy Labs runs on Lucy. Every internal workflow, every coaching session, every agent we ship; if it doesn't work for us, it doesn't ship to anyone else.
Lucy Labs runs on Lucy. This is proof, not theory. The company is a system of agents running the operating model we sell.
Today, Travis and an agentic team compress the back-office work that would otherwise need a 5–10-person team. The product, the platform code, the marketing content, and the legal/compliance scaffolding were built by one person + AI in months.
The point isn't that Lucy Labs has solved company-building with AI; the point is that the scaffolding, framework, and operating discipline already exist, and every new hire ramps into a system that amplifies their work instead of slowing it down.
The org chart is the operating model.
The cultural fits (direct communication, specific over abstract, bias toward shipping, comfort with truth over comfort, humanistic thinking, "we use what we build" instinct) are detailed in the open-roles section below. The full Principles + Values + Guardrails taxonomy lives on how we operate.
The same trust posture Lucy ships to customers is the trust posture inside the company. No surveillance of customers, teammates, or ourselves. No compromising principles for revenue. No selling Lucy where the architecture refuses what the customer expects. See the refusal scene.
Operating model. Principles are how we build. Values are how we behave. Guardrails are what we refuse.
What you build at Lucy Labs is yours. The skills, the prompts, the workflows, the agents — they stay with you when you leave, and they compound across the next decade of AI.
Portability
What you build is yours, and it leaves with you. The skills, prompts, workflows, and agents you build with Lucy survive role changes, employer changes, and the next model generation.
Compounding
Universal AI skills compound across products, providers, and model generations — the patterns you learn this year still work next year, getting sharper, not obsolete.
GP-001. AI augments human capability; AI does not replace human judgment.
GP-001 applies to the team, not just the customer. The work you do here augments your career; the company is structurally committed to the same posture for its own people that it commits to for customers. Career portability is structural — capability stays with the human who built it (GP-003). See what we believe about people for the longer treatment.
Lucy Labs is in design-partner development — we are a lab, and that means always discovering, learning, and growing. The people we hire now help shape what Lucy Labs becomes.
Real product, real customers, real roadmap. Pre-GA. Post-prototype. The product, the trust architecture, and the coaching loop are available. Some surfaces and controls are planned. Specific certifications (SOC 2 Type II, ISO 27001, ENS HIGH) and the manager-side admin surface are planned for GA. The labels are honest; the roadmap is real.
See the trust architecture for the full stage-honest item list.
What "design-partner stage" means for a candidate
It means the work you do shapes the company that emerges, not the work you do operates a company that already exists. It means the operating motion changes month over month as we learn. It means decisions you make at this stage compound — your fingerprints are on the architecture for years. It means the comp / equity / time mix is shaped by the stage; we are transparent about that in the conversation.
What we are NOT yet
We are not at general availability. We are not a multi-tenant enterprise platform. We are not a Series-stage company with a fully-staffed function for every problem; the people we hire now are people who can build the function as they do the work. The honest signal about stage is part of the candidate-honesty contract.
Open roles, how to apply, and what to expect from the conversation.
Lucy Labs hires for slots, not for ladders. The roles below are the slots open right now — what we'll need someone to do, the level of context required, the kind of person we think will succeed in the role.
First wave
The Work Modeling team
Ten specialists build the loop from captured evidence to work understanding to coaching.
Engineering Lead / Staff Product Architect
Owns technical coherence across every Lucy system, hands on.
Desktop / Client Capture Engineer
Builds the always-on desktop agent on Windows and macOS.
Senior Backend / Data Platform Engineer
Owns evidence ingestion and the lifecycle backbone under everything else.
Privacy / Security / Trust Engineer
Builds the controls that make Lucy safe to use with real work.
Staff Applied Behavior Inference Engineer
Turns messy work evidence into actions, tasks, goals and workflows.
Data / Evaluation Engineer
Builds the ground truth that keeps Lucy from being confidently wrong.
Knowledge Graph / Ontology Engineer
Owns the task, tool, skill and workflow model Lucy reasons over.
Product / Full-Stack Engineer, Coaching and Review UX
Makes what Lucy infers correctable, explainable and worth trusting.
AI Platform / MLOps Engineer
Owns model routing, versioning, eval execution and AI observability.
QA / Release / Test Automation Engineer
Builds the quality system, not a manual test pass.
Broader and scale roles
The engineering organization around the first wave
Six open roles cover the broader product stack and the organization that grows around it.
Backend Engineer
Builds the core services and APIs the product runs on.
Frontend Engineer
Owns the interface and interaction layer of the Lucy coach.
Platform / DevOps Engineer
Owns infrastructure, pipelines and the reliability everyone ships on.
ML / AI Engineer
The broad AI role for the founding phase, before it splits.
Engineering Manager
Player-coach for a team of five to eight engineers.
VP Engineering
Owns the whole engineering organisation, reporting to the CEO.
What we look for — cultural fits
- Direct communication. Specific over abstract. Comfort with truth over comfort.
- Bias toward shipping. Comfort with non-final answers as you ship them.
- Coach posture for customers and for teammates — we coach, we don't act on people's behalf.
- Humanistic thinking — the person in the work is the point.
- "We use what we build" instinct — if you wouldn't use it, don't ship it.
What we don't look for — cultural unfits (politely)
- Surveillance posture, dashboard-fixation, or measurement-of-people-rather-than-systems instincts.
- "Move fast and break things" without the "fix what we broke" follow-up.
- Hype-laden communication style ("transform," "synergize," "leverage" as filler).
- Comfort hiding behind process instead of taking decisions.
What to expect from the conversation
- A first conversation with the founder or a senior team member (45 minutes; conversational, not stagecraft).
- A working session on a problem from the actual operating context (90 minutes; we look at how you think and how you ship).
- A reference / verification round (we talk to people who've worked with you).
- An offer or a clean no.
Frequently asked questions
Is Lucy Labs hiring?
Yes. Lucy Labs is in design-partner development and is hiring for the open roles listed above. Applications start with the contact form.
Where is Lucy Labs based?
Lucy Labs is based in Madrid, Spain, with a Delaware C-Corp parent and a Spanish operating subsidiary. Every role currently listed is based in Madrid; the exact working arrangement is confirmed during the application conversation. The dual-entity structure is described on /about.
What stage is Lucy Labs at?
Lucy Labs is in design-partner development: pre-general-availability and post-prototype. The product, trust architecture, and coaching loop are available; some surfaces are planned; some certifications are planned for GA. The full stage-honest roadmap is on /trust.
What's it like to work at Lucy Labs?
Lucy Labs is a small, high-pace team built around direct communication and a bias toward shipping. We use what we build: Lucy Labs runs on Lucy. The cultural fits, unfits, principles, values, and guardrails are described above and on /about.
Does Lucy Labs sponsor visas?
Lucy Labs evaluates visa sponsorship case by case. Raise it during the application conversation.
What does the interview process look like?
Lucy Labs starts with a 45-minute conversation, followed by a 90-minute working session on a real operating problem. A reference round follows, then an offer or a clean no. The full process is in the open-roles section above.
