Open Government Products

Over two years, our citizen testers did something most of us never think about when we drive: they enrolled their own vehicle plates, drove through our cameras on their usual routes, and told us every time something went wrong. That feedback was the experiment. Without it, ERP X would have stayed an idea OGP believed in, rather than one we could prove. To our 403 citizen testers, once again: thank you for driving this with us 🙏🚗 If you would like to help test what we build next, join us at https://lnkd.in/gJbjyMxx

ERP X started with a question: is there a different way to do road pricing in Singapore? 📸 We could have left it as a hunch. Instead we spent two years testing it on real roads, with real drivers, before deciding anything. What we found is worth sharing openly, not because it is launching, but because of how we got there: prototype early, test cheaply, bring citizens in from the start, and let the evidence decide. Swipe through for the story. Full report at https://lnkd.in/gS4jqTS3 🔗

ERP X has reached the finish line. 🏁🏁 For two years, 403 citizen testers joined us in an experiment to test a camera-based alternative to ERP. This was not only a test of new technology. It was also a way to show that government can trial national infrastructure quickly and affordably before committing to build it. While there are no plans to roll out camera-based road pricing, it leaves behind findings that can shape future conversations on Singapore's transport policy, and a way of working that other Public Service teams can adopt. Read the full ERP X report: https://lnkd.in/g_hVvfyH

Good government is not only about getting the answer right. It is about being willing to test whether there is a better way, and being honest about what you find. For two years, 403 citizens drove around Singapore as they always had. What changed was on the roads around them: a network of cameras quietly read their licence plates as they passed, and an app "charged" them for ERP each time. 𝐓𝐡𝐢𝐬 𝐰𝐚𝐬 𝐄𝐑𝐏 𝐗, 𝐎𝐆𝐏'𝐬 𝐞𝐱𝐩𝐞𝐫𝐢𝐦𝐞𝐧𝐭 𝐭𝐡𝐚𝐭 𝐞𝐱𝐩𝐥𝐨𝐫𝐞𝐬 𝐜𝐚𝐦𝐞𝐫𝐚-𝐛𝐚𝐬𝐞𝐝 𝐚𝐥𝐭𝐞𝐫𝐧𝐚𝐭𝐢𝐯𝐞𝐬 𝐭𝐨 𝐄𝐑𝐏. 𝐓𝐨𝐝𝐚𝐲, 𝐰𝐞 𝐚𝐫𝐞 𝐩𝐮𝐛𝐥𝐢𝐬𝐡𝐢𝐧𝐠 𝐰𝐡𝐚𝐭 𝐰𝐞 𝐥𝐞𝐚𝐫𝐧𝐭. ERP X was never meant to replace ERP 2.0. Nonetheless, our findings are a useful starting point that future camera-based solutions can build on, and can help shape the future of transport policy. More than just a proof of technical viability, ERP X is also proof of a way of working. It shows that government can test national infrastructure quickly and affordably before committing to build it. Instead of committing to a large build first and fixing things later, the team prototyped early, tested with real users, and refined as they learnt. This approach is familiar in software development, and ERP X shows it works for national infrastructure too. It's a practice we believe more teams across the Public Service can adopt. Singapore has never shied away from trying things others said could not be done. ERP X is a small reminder that we can keep doing that: asking whether there is a better way, and pushing forward. Read the full report here: https://lnkd.in/g4yV7iZc

What makes innovation actually work in the public sector? At the CPF Board Learning Festival, OGP’s Director Hongyi Li shared what years of building public products has taught him about the people, systems and processes that make or break a new idea. Swipe through to see what that looks like in practice 👉

People picture government tech as slow, top-down, permission-first. That's not the OGP our engineers describe. When AI started reshaping how software gets built, our teams didn't wait for an official policy, they tried the tools on real problems and shared what worked. We asked Si Mun, a senior software engineer on our Social360 team, working to help give social workers back time with the people they're actually there for, to tell it in her own words: “The way I think about AI at work shifted when my tech lead told me about 'superpowers', a skills framework for coding agents. He was ahead of the curve, experimenting with it before Claude Opus existed, before AI agents were all anyone could talk about, and he wanted us to learn it too. But the bigger takeaway was this: My superpower is the smart, competent, and insanely motivated people I get to work with. Across every team I've been on, one thing held: no one waits for a memo. When AI started changing how we work, my team just dove in. People tested tools on real problems and shared what worked almost as fast as they learned it. One teammate compared orchestrators for running coding agents and shared what he found with a few of us. No polished deck, just 'here's what I tried, here's what worked.' Senior managers were digging into the same tools, and they set aside time on Fridays for exactly this kind of sharing, across the org and within our healthtech teams. I've used that dedicated learning time too. Ahead of a security review, my team found an AI vulnerability scanner that could surface issues before a pentest would. I ran it on real code and shared back what was useful, so the next person wouldn't start from zero. None of this started as a mandate. It grew bottom-up until the org made space for it. That's the culture: smart, motivated people trying things and building on each other." We're hiring senior engineers who want to work this way. Come build with us: https://lnkd.in/guq5yky4

Meet the team behind Polyglot (https://lnkd.in/gEaTCMqv), an AI-powered tool translating government information into Chinese, Malay and Tamil. Harish Venkatesan and Zhong Jun Hsu helped build it, and they sat down with NUS Computing to share how their very different paths led them to public service. Read more below 👇

Building tech for public good isn't limited to Singapore. Neither are the challenges that come with it: product thinking, data sharing, and getting products from prototype to production. This month a few of us were in Bangkok for BDI - Big Data Institute (Public Organization)'s Open Data Hackathon. We shared what we've learned building products like a startup inside government, so teams in similar organisations across Southeast Asia can do the same. A few of the things we shared: 💡 Most hackathon products fail before demo day. Not because the product doesn't work, but because no one knows whether people actually need it. 💡 Start absurdly small, with someone who has real skin in the game, before you try to convince anyone at scale. 💡 To decide whether something is worth building, ask three questions. Opportunity: is this actually a big enough problem? Velocity: how fast can you get an honest signal back? Traction: is anyone pulling this out of your hands? If you're interested in making an impact in Singapore and beyond, we're hiring: https://lnkd.in/g7JFJJAr

Real change scales when public officers do not just solve their own problems, but help others solve theirs too. That was the idea behind the Community Mini-Sprint 2026, a peer-led programme for public officers. Over two intensive days, six teams from SG Courts, National Cancer Centre Singapore, National Council of Social Service (NCSS), SingHealth, Singapore Food Agency and Singapore Prison Service and worked on real workplace challenges using no-code and low-code whole-of-government tools. Made up mainly of officers in non-technical roles, the teams refined their problem statements, mapped opportunities and blockers, tested early ideas, defined success metrics, and began building practical solutions. Throughout the sprint, they were guided by peer facilitators from across the Public Service. These facilitators drew on their own experience improving workflows in government to challenge assumptions, ask better questions and help teams move from ideas to action. A big thank you to Benjamin Mok, Benjamin Pang, Fang Jie Kwok, Javier Ang, Jerry Loo, Sally Sia, Xinnie Gan and Tan Teck Wei Victor for stepping up and paying it forward, and to Natalie Mae Tan, Senior Product Designer at OGP, for equipping them with the skills to guide the teams effectively. The sprint does not stop here. Over the next few weeks, teams will continue refining their ideas through office hours, with support from OGP and peers across the Public Service. See how far they have come at the Community Mini-Sprint 2026 Showcase on 5 August 2026. Hear directly from the teams about the problems they tackled, the solutions they built and what they learnt along the way. Open to public officers only. Join the community and register by 29 July 2026: https://lnkd.in/gR-ZhQy8

This morning, Andrew (Yi Rong) Tan and Subeesh Basheer joined the Flying Dutchman and Lavinia live on OneFM 91.3 🎙️ The half-hour chat covered two of our Limited Releases: PlayTime (https://playtime.gov.sg/), which helps parents find family-friendly activities nearby with three simple questions, and Undercover (https://undercover.gov.sg/), which plots cooler walking routes by factoring in shelter, sun position and building shadows. They also shared how these came about: both were built at this year's Hack for Public Good, OGP's annual hackathon, and shared publicly so anyone can try them and tell us what to improve. Catch the full segment here: https://lnkd.in/gavERaJ2