Singapore's pipeline from research to spinout is getting bigger. The National Graduate Research Innovation Programme (National GRIP) began in January 2025. It is backed by S$50 million in financial and in-kind support over five years, aims to train up to 300 teams by 2028 and targets more than 150 spinouts by 2030. It builds on the NUS GRIP programme, which has funded teams since 2018, with up to S$250K available for high-potential teams.
These programmes do what they are designed to do. A team leaves with validated science, often an IP licence from the university, and the discipline of having tested a commercial hypothesis. What the team usually leaves without is everything a Series A investor will look for next.
What the grant does not cover
| The team usually has | The company still needs |
|---|---|
| Validated science and a licence | A product that runs outside the lab, every day |
| A prototype or pilot rig | Production software, data pipelines and security |
| Deep technical credibility | Evidence a customer or regulator can read |
| A research team | A company: positioning, brand, a commercial model |
| Grant reporting experience | An investor story with milestones and financial logic |
This is the difficult middle. The science is proven enough to be exciting and not yet productised enough to be fundable. Hiring an engineering team takes months. An agency can build software, but it will not make the architecture decisions that a regulated product needs.
One team that crossed it
WaterDoctor was founded by NUS PhD researchers in environmental biotechnology, microbial ecology and AI. It builds AI-integrated biofilm reactors for water treatment and aquaculture. The science was strong. The company needed a production system that could learn a noisy biological process and show a regulator what it was doing.
Since 2024, wGrow has built and operated that system: a crew of four narrow agents (ingestor, controller, reporter and eval) and two senior engineers who set the safety limits and sign every report that goes to a regulator. WaterDoctor reports that its pilot tanks cut pollutant discharge by more than 50%, water exchange by 90% and energy per unit by 30%, compared with the prior baseline. The company has completed its Pre-A round and is now in Series A conversations.
What we learned
- Scientists keep the science. WaterDoctor's process scientists own the biology. The engineers own the agent stack and the audit trail. Mixing the two roles slows both down.
- Lab reality differs from sensor reality. Lab assay results arrive days after the sensor reading. A controller that assumes they line up will make confident mistakes. The fix was a delayed-supervision channel.
- Hard limits before autonomy. When the system tuned aeration too aggressively, engineer-written guardrails caught it. Changes of more than 15% from the rolling baseline now need human approval.
- Evidence is the product. For a deep-tech company, an operating record the regulator can read is worth as much as the reactor itself when it comes to raising the next round.
If this is where you are
If you have finished a grant programme, hold a licence and have a pilot that works, and you now need a product, a company and an investor story, that is the stage wGrow Labs is built for. Please do not send unpublished IP in a first message. A short description of the problem and what exists today is enough.
- NUS News: National GRIP launch · nationalgrip.sg
- NUS GRIP programme
- wGrow: WaterDoctor engagement write-up
- Outcome figures are as reported by WaterDoctor in wGrow's case study