Search for a deep tech venture studio in Southeast Asia and you get a data page, three press releases and an opinion column. That result is itself the finding. The region does not lack science, and lately it does not even lack money. What it lacks is the layer between the two: builders who take a technology that works in a lab and turn it into a company that runs hardware in the field, in markets where the customer is a factory in Java or a hotel in Bali, not a cap table in Singapore.
We build that layer. EX EPIC finances, patents and deploys deep tech from a base in Bali, so this is a view from inside the gap, not a directory scraped from outside it. Here is what the numbers say, which builder models actually exist in the region, and the questions that separate a studio from a programme with a studio's name.
Why Southeast Asia is short of deep tech builders, not deep tech
Start with the size of the field. Tracxn counts 2.68K deep tech companies in Southeast Asia, of which 795 are funded, 200 have reached Series A or later and 4 are unicorns, with $9.77B raised to date. The same source puts 2025 at $562M across 53 rounds, against $540M across 118 rounds in 2024.
Read that twice. Slightly more money, less than half the rounds. The capital did not leave deep tech. It stopped spreading.
Money into fewer hands
The whole regional market shows the same shape. Between January and July 2026, Southeast Asian startups raised $12.8bn, up 137% year on year, while equity rounds fell 30% from 255 to 178 and the average round rose from $21m to $72m. The report's own verdict is blunt: it is the same market with the money pushed into fewer hands, and a good share of it went into data centres and AI infrastructure rather than operating startups.
For a deep tech founder this is the key fact. A market of fewer, larger rounds rewards companies that are already de-risked. It punishes the stage where a technology still needs a pilot plant, a patent strategy and a first paying customer. That stage is exactly where a venture studio is supposed to work.
Singapore books the capital, the region runs the operations
The second structural fact is geography. Singapore took more than 60% of the region's deal count in 2025, and the same report warns that domicile is not operations: many Singapore rounds fund assets and businesses elsewhere. Singapore itself is leaning in. Deep tech's share of Singapore venture deal value rose from 11.1% in 2022 to 24.7% in 2025, backed by S$37 billion under RIE 2030 and S$1 billion for the Startup SG Equity scheme.
So the region splits in two. The cap table lives in Singapore. The biomass, the waste streams, the energy demand and the factories live in Indonesia, Vietnam, Malaysia and the Philippines. A builder that only operates in the first half cannot deploy into the second, and deep tech that is never deployed never earns.
What a deep tech venture studio has to do that a fund cannot
The case for studios in deep tech is well made. Unlike software, deep tech startups face long R&D cycles, capital-intensive prototyping, regulatory complexity and fragmented value chains, and traditional venture capital struggles with those time horizons. A fund writes a cheque and waits. A studio pays for the years before there is anything to invest in: it files the patents, recruits the operators and runs the first deployment.
We have written separately about what a deep tech venture builder actually does and how the instrument compares under venture studio versus venture capital. In Southeast Asia, three jobs matter more than anywhere else:
- Cross-border structuring. Money raised in one jurisdiction, IP filed in several, hardware installed in another. Someone has to hold that together before a CFO exists.
- Operator supply. The scarce resource is not scientists. It is people who can install, maintain and sell a physical system in an emerging market.
- Deployment before scale. In a market of fewer, bigger rounds, the first units in the field are what turn a science project into a fundable company.
The four builder models operating in the region
Most of what calls itself a deep tech venture studio in Southeast Asia falls into one of four models. They are not ranked. They do different jobs, and founders get hurt when they pick one expecting another.
State-backed deep tech investors
Singapore's answer is SGInnovate, which was set up to help people build technology-intensive products born out of scientific research, working on human capital and investment capital. It is patient, well connected and deliberately national. It is an investor and ecosystem builder, not a company creator, and its gravity is Singapore.
Accelerator programmes that call themselves labs
Malaysia's Startup Deep Tech Ventures Lab, run by Sunway iLabs and Cradle, is a six-month programme that selected 20 startups from 69 applicants. In Singapore, Moringa Ventures describes itself as a deep tech venture builder established in 2025, and its first programme, Momentum, is 10 weeks long with up to S$250,000 in funding for pre-incorporation and early-stage teams.
Both are useful. Both are cohort programmes: the team arrives with the technology, gets structure and access, and leaves. That is the accelerator model, whatever the label says. Founders should know which one they are signing up to.
AI-native and climate studios
The Radical Fund, a Southeast Asian climate VC, launched The Studio with a 12+ person team and a promise to take ventures from concept to Series A in under 24 months, sitting alongside a fund that has invested in 10 companies across Singapore, Indonesia, Malaysia and Vietnam since 2023. This is a real studio model: it co-builds rather than just advises. Its centre of gravity is software and AI infrastructure applied to climate and industry.
Operator builders with deployed hardware
The fourth model is the rarest in the region: a builder that owns the whole chain from patent to installed unit. This is where EX EPIC sits. Our record, published on exventure.co, is a EUR 160M+ capital track record across 4 continents, Zero-X waste-to-energy systems with 200+ units deployed across 11 countries, 24 patent families filed through EX IX with around 100 more validated in the pipeline, and 250+ operators trained and deployed through EX EPIC Academy.
It is the same logic that shaped deep tech venture builders in Europe, with one difference. In Europe, builders substitute for a broken financing chain. In Southeast Asia, they also have to substitute for missing operators.
How to vet a deep tech venture studio in Southeast Asia
Press releases all sound alike. These five questions do not have press release answers.
1. How many units are running in the field, and where? Not pilots announced. Units installed, with the country named. If the answer is zero, you are talking to an accelerator or a fund.
2. How many patent families has the studio filed, and who owns them? A deep tech company without protected IP is a consulting project. Ask whether the IP sits with the venture, the studio or the university, and whether you can read the assignment.
3. Who operates the technology after the demo day? Ask for the names of the people who will install and maintain the system. A studio without an operator pipeline will hand you a deck and a hiring problem.
4. Where is the money booked, and where does the hardware run? Capital in Singapore and deployment in Indonesia is normal. A studio that cannot explain how it bridges the two has not done it yet.
5. What does the studio take, and for what? Studios take large founding stakes because they do the founding work. That is fair only if questions one to four have strong answers. If the studio does not deploy, patent or operate, the equity is paying for advice.
Run EX EPIC through the same five and hold us to it: 200+ units in 11 countries, 24 patent families, 250+ operators, a UK regulated financial arm in the FCA-authorised EX FI. If a studio you are talking to scores better, go with them. The point of the test is that it works on everyone.
Where EX EPIC fits, from Bali
EX EPIC was founded and is led by Julien Uhlig, and the ecosystem runs from Bali. That is not a lifestyle choice. It puts the builder next to the deployment markets: tropical biomass, waste streams, off-grid energy demand and a talent pipeline that trains operators on real systems.
For a researcher sitting on a lab-stage technology, the route in is usually a spinout. If that is your stage, start with how to spin out a company from a university. If you are an investor sizing the risk, read why deep tech startups fail: nearly every failure mode on that list is one a studio exists to absorb.
Southeast Asia will not run short of deep tech capital in the next cycle. It will run short of companies ready to take it. Builders that deploy are how that changes.
FAQ
How many deep tech startups are there in Southeast Asia?
Tracxn counts 2.68K deep tech companies in the region as of early 2026. Only 795 have raised funding and 200 have reached Series A or later, which shows how steep the drop off is after founding.
Is a venture studio the same as an accelerator?
No. An accelerator runs a fixed cohort of teams that already exist, for weeks or months, then lets them go. A venture studio creates the company, holds a founding stake and keeps operating it. Several Southeast Asian programmes use studio language for what is structurally an accelerator.
Does Singapore fund deep tech startups directly?
Yes. SGInnovate is Singapore's deep tech investor and talent builder, and the government has committed S$37 billion under RIE 2030 plus S$1 billion for the Startup SG Equity co-investment scheme.
Can a deep tech startup be based in Indonesia rather than Singapore?
It can operate there, and for hardware tied to local biomass, waste or energy demand it often should. Many teams still incorporate or raise through Singapore, because that is where most regional deals are booked. The builder's job is to make both halves work together.