Blab AI
AngelHack Uses Internal Hackathons to Bridge AI Adoption Gap in Enterprises
AngelHack, a Singapore‑based innovation firm, is turning internal hackathons into a practical solution for companies that struggle to translate large language models (LLMs) into business‑relevant applications. In a recent interview with Computer Weekly, Ryan Chew, director of AngelHack DevLabs, explained that while AI adoption is accelerating, many organisations still lack the know‑how to build AI tools that fit their specific workflows.
Chew’s insight is straightforward: instead of hiring external consultants who may not understand a company’s day‑to‑day operations, AngelHack runs week‑long hackathons that train non‑technical staff—such as marketing, finance, and operations teams—to create AI‑powered prototypes. He said, "Many companies like to get consultants from external parties who say, ‘Hey, try to do AI for me.’ But they don’t know your workflow, they don’t have privileged information, and they don’t exist day‑to‑day in operations… You just need to spend a week to teach the people who are the actual stakeholders, and they can produce much better results because they know their workflows best." The approach has proven effective for a range of clients, from local telco Singtel to the government‑linked Defence Science and Technology Agency BrainHack, which uses hackathons to identify and recruit top tech talent.
Beyond training, AngelHack also organizes large‑scale external hackathons for tech giants such as Microsoft, helping them adopt tools like Copilot. The firm’s global developer database enables it to host events that attract participants from around the world, and the resulting prototypes often move into the market. In Singapore, AngelHack is collaborating with Digital Industry Singapore (DISG) and Nvidia to run accelerator programmes that transform prototypes into scalable businesses.
For organisations that lack in‑house engineering expertise, AngelHack’s DevLabs arm can build minimum viable products (MVPs) and custom software applications. The service is positioned as a bridge that eases the transition into frontier technologies such as AI and blockchain.
The rise of LLMs also raises organisational questions. Chew describes the issue as a "surface area" problem: if a single developer using AI can now perform the work that previously required a team of five, companies must rethink how they structure engineering teams. "Do you have enough features for everyone to work on? And if you do, you’d need to redesign your organisational structure within your engineering team so that their roles are clear, whether that’s in auditing or reviewing pull requests," Chew said. He argues that AI should be a multiplier, not a cost‑cutting tool. "If you have AI empowering engineers, you should be able to produce more and not try to cut people," he added.
Governance, risk and compliance remain key concerns. Chew notes that "No AI is 100% correct, and no human is 100% correct. I can hold humans accountable, but I cannot hold AI accountable." Consequently, AngelHack’s current focus is on internal workflows where the risk profile is lower than in external‑facing applications.
AngelHack’s workforce is about 70 employees, and demand for its academy, hackathon and custom solutions businesses is rising. Chew highlighted that small and mid‑size enterprises (SMEs) are increasingly engaging DevLabs to understand how AI can improve internal efficiencies or be integrated into existing products.
Geographically, AngelHack has expanded beyond Singapore. The firm has organised hackathons for customers in Australia and Brazil and plans a multi‑city hackathon across the United States later this year.
The company’s growth reflects a broader trend: as the barrier to technology development falls, the appetite for expert facilitation remains strong. "Even if you know how to cook, if cooking is made super easy, sometimes you just prefer to eat at a restaurant," Chew quipped.
In summary, AngelHack is positioning internal hackathons as a low‑risk, high‑impact method for enterprises to experiment with LLMs, train staff, and develop market‑ready AI solutions. The firm’s partnerships with government agencies, industry giants and semiconductor leaders underscore the commercial and strategic value of this approach. As AI continues to reshape software engineering and organisational structures, the role of facilitation and internal training may become increasingly central to successful adoption.
Chew’s insight is straightforward: instead of hiring external consultants who may not understand a company’s day‑to‑day operations, AngelHack runs week‑long hackathons that train non‑technical staff—such as marketing, finance, and operations teams—to create AI‑powered prototypes. He said, "Many companies like to get consultants from external parties who say, ‘Hey, try to do AI for me.’ But they don’t know your workflow, they don’t have privileged information, and they don’t exist day‑to‑day in operations… You just need to spend a week to teach the people who are the actual stakeholders, and they can produce much better results because they know their workflows best." The approach has proven effective for a range of clients, from local telco Singtel to the government‑linked Defence Science and Technology Agency BrainHack, which uses hackathons to identify and recruit top tech talent.
Beyond training, AngelHack also organizes large‑scale external hackathons for tech giants such as Microsoft, helping them adopt tools like Copilot. The firm’s global developer database enables it to host events that attract participants from around the world, and the resulting prototypes often move into the market. In Singapore, AngelHack is collaborating with Digital Industry Singapore (DISG) and Nvidia to run accelerator programmes that transform prototypes into scalable businesses.
For organisations that lack in‑house engineering expertise, AngelHack’s DevLabs arm can build minimum viable products (MVPs) and custom software applications. The service is positioned as a bridge that eases the transition into frontier technologies such as AI and blockchain.
The rise of LLMs also raises organisational questions. Chew describes the issue as a "surface area" problem: if a single developer using AI can now perform the work that previously required a team of five, companies must rethink how they structure engineering teams. "Do you have enough features for everyone to work on? And if you do, you’d need to redesign your organisational structure within your engineering team so that their roles are clear, whether that’s in auditing or reviewing pull requests," Chew said. He argues that AI should be a multiplier, not a cost‑cutting tool. "If you have AI empowering engineers, you should be able to produce more and not try to cut people," he added.
Governance, risk and compliance remain key concerns. Chew notes that "No AI is 100% correct, and no human is 100% correct. I can hold humans accountable, but I cannot hold AI accountable." Consequently, AngelHack’s current focus is on internal workflows where the risk profile is lower than in external‑facing applications.
AngelHack’s workforce is about 70 employees, and demand for its academy, hackathon and custom solutions businesses is rising. Chew highlighted that small and mid‑size enterprises (SMEs) are increasingly engaging DevLabs to understand how AI can improve internal efficiencies or be integrated into existing products.
Geographically, AngelHack has expanded beyond Singapore. The firm has organised hackathons for customers in Australia and Brazil and plans a multi‑city hackathon across the United States later this year.
The company’s growth reflects a broader trend: as the barrier to technology development falls, the appetite for expert facilitation remains strong. "Even if you know how to cook, if cooking is made super easy, sometimes you just prefer to eat at a restaurant," Chew quipped.
In summary, AngelHack is positioning internal hackathons as a low‑risk, high‑impact method for enterprises to experiment with LLMs, train staff, and develop market‑ready AI solutions. The firm’s partnerships with government agencies, industry giants and semiconductor leaders underscore the commercial and strategic value of this approach. As AI continues to reshape software engineering and organisational structures, the role of facilitation and internal training may become increasingly central to successful adoption.
