We build what the market needs, not what the hype demands.
Skaler is an operator-led venture studio in Lisbon. We identify structural gaps in traditional industries, validate that demand exists, and build companies to fill them. We prove the business works first. Then we bring in capital to make it bigger. We do not chase trends. We chase durable businesses.
Every venture starts with a gap, not an idea.
We do not brainstorm in a vacuum. We study the market until we find a problem that is structural, expensive, and currently unsolved. Then we validate that people will pay for the solution before we write code.
- 01
Map the stakeholders.
Buyers, sellers, intermediaries, regulators, institutions. What do they need? What do they lack? Where does money leak?
- 02
Quantify the pain.
How much time is wasted? How much risk is hidden? How many transactions fail because of information asymmetry? If the numbers are not large, we move on.
- 03
Check for traction signals.
Are people already trying to solve this with spreadsheets, informal networks, or expensive consultants? That is evidence of demand.
- 04
Validate willingness to pay.
We talk to potential customers before we build. Not surveys. Conversations. If they will not pay for a prototype, we kill the idea.
- 05
Model the unit economics.
Does the math work at realistic volumes? If it only works at unicorn scale, it is not for us.
Only after these five steps do we commit capital and team. This is why we have launched three ventures with zero write-downs to date. Not because we are lucky. Because we only build where the gap is real and the economics are clear.
This process produced our three ventures. InspectOS answers the verification failure: no standards, hidden defects, and a wide gap between paper records and physical reality. HomeOS answers the information failure: fragmented registries and buyers making the largest decision of their lives without reliable data. Rezerva answers the transaction failure: a noisy public market and an off-market segment with no professional infrastructure. Each venture maps to a specific, measured gap.
We prove first. Then we partner.
We fund early stages ourselves. This gives us speed, control, and the discipline to say no to bad opportunities. When a venture reaches growth stage with proven traction and unit economics, we partner with patient capital to accelerate what already works. We do not raise for ideas. We raise for scale.
Why self-fund the early stage
Speed
No fundraising cycles. No pitch decks to investors who do not understand the market. We build while others are still meeting.
Control
We set the strategy, hire the team, and own the timeline. No external pressure to pivot for fashion.
Discipline
We validate before we spend. If the idea does not survive customer conversations, we kill it cheaply and move on. No capital deployed until demand is proven.
Alignment
When we do bring in partners, they enter a venture that already has customers, revenue, and a working model. Their capital de-risks growth, not discovery.
How we partner
Venture level first
Partners invest in specific ventures with defined governance and transparent reporting. Studio-level capital would be considered only for long-horizon infrastructure, such as team housing and offices, and only from patient, strategic partners. We do not take quick-win capital.
Patient capital preferred
Family offices, asset managers, strategic operators. People who measure in decades, not quarters.
Structured for alignment
Equity participation, board or observer seats, quarterly reporting with agreed KPIs. Our partners know exactly what they own.
Our model is simple. We prove the business works on our own capital. Then we bring in partners to make it bigger. Their money accelerates proven traction. It does not subsidize experiments.
Portuguese real estate is our beachhead. It is not our boundary.
We started in Portuguese real estate because the gaps were obvious, the regulatory tailwinds were strong, and our local expertise gave us an unfair advantage. But the system we built, the Company OS, the AI-native operating model, the compounding data layer, is designed to travel.
What we look for in a new market
Fragmented data
Industries where information is trapped in silos, paper records, or informal networks.
Weak or absent standards
Markets where verification is not mandated but the risk is real. Whoever builds the verification layer sets the standard.
Institutional pain
Banks, insurers, asset managers, or regulators who need better data to price risk, prevent losses, or enforce rules.
Local expertise gap
A market where deep domain knowledge is scarce and expensive, but where our systems-thinking approach can close the gap quickly.
Current portfolio
InspectOS (property inspections and condition verification) · HomeOS (buyer intelligence and registry cross-referencing) · Rezerva (off-market transaction network), all Portuguese real estate.
Evaluating next
Insurance risk modeling (data layer extension) · Construction compliance software (regulatory extension) · Property management automation (customer extension). Same country, new industries.
Future horizon
Spain and Italy (similar fragmentation, similar institutional pain; Company OS and talent systems are portable, local expertise gets rebuilt through partnerships). Beyond that, other traditional industries where physical assets, fragmented data, and institutional risk converge: construction, logistics, agriculture, energy retrofit. The same pattern repeats wherever verified condition data does not exist.
We are not a real estate company. We are a systems company that currently operates in real estate. The model travels where the conditions are right.
One system. Multiple ventures. Compounding leverage.
The Company OS is the infrastructure behind every venture we build. It is not a product. It is how we operate. Research, recruitment, product standards, marketing, sales, and capital allocation, all running on shared systems that improve with every launch.
Research and monitoring
Market, competitor, regulatory, demographic, and economic intelligence, continuously updated.
Recruitment and talent systems
University pipelines, proof-of-work screening, structured development through weekly workshops and monthly sprints.
Product and engineering standards
Shared components, quality gates, reusable infrastructure.
Marketing, sales, and distribution
SEO systems, outreach playbooks, partnership frameworks.
Capital allocation
Financial modeling, staged deployment, disciplined kill criteria.
Performance data from each venture feeds back into the system. Every new company starts smarter than the last. We measure the OS by outcomes: how fast we launch, and how little capital it takes.
The OS is designed to adapt. The research module works for any traditional industry. The recruitment module sources for any technical domain. The product standards apply to any data-heavy venture. This is why we can enter new markets faster than competitors who build from scratch. Deep dive lives on the Studio page.
AI expands our surface area. Humans control the outcomes.
We run on an AI-native operating system. Narrow agents execute defined jobs. A shared data layer compounds with every interaction. People own every judgment that carries liability. This is not a feature we added. It is how the company works.
Shared context layer
The Property Condition Graph: inspection records, registry cross-references, transaction outcomes, regulatory rules. Machine-readable, so agents can reason over it. Every venture reads from it and writes back to it.
Agent fleet, narrow jobs
Registry and due diligence agents. Report drafting and QA agents. Regulatory monitoring and lead scoring agents. Matching agents for off-market deal flow. Every agent has a narrow job, a measurable output, and a human approval path.
Human judgment layer
People own strategy, compliance sign-off, client relationships, and every certified output. No certified report ships without a qualified professional signing it.
The engine room
A dedicated AI systems team scans new developments, prototypes against real venture problems, and certifies what works before it spreads. Weekly AI tooling sprints and monthly in-house workshops train the whole team on certified workflows. Failures become reusable rules.
Measurement
We track AI contribution like any other input: share of report sections auto-drafted, time from inspection to report, accuracy against human baselines, hours saved per workflow. What does not measurably help gets removed.
Governance
Approval paths, risk thresholds, escalation rules, and EU AI Act readiness are built into the system. Where outputs carry certification and liability, governance is not overhead. It is the license to operate.