Senior Software Engineer, AI Native

Who We Are

Open Government Products (OGP) operates like a startup within the government; We are a fast-paced, dynamic team of engineers, designers, and product managers dedicated to public good. We autonomously identify opportunities to solve public problems using tech, rapidly prototype, and launch solutions — from citizen apps to automating public agency operations. Embracing an open-sourced and flat hierarchy, we cut through bureaucracy to focus on real-world, user-centric innovations that drive change.

The Team

We're hiring for the AI Native team. Our mandate: rethink how product development happens across OGP, with AI at the centre of how people work.

We work with product teams and with OGP's central tooling team. We build shared AI tooling on top of existing infrastructure and help product teams adopt it across how they code, plan, and operate. Our success metric: teams across OGP move faster because of how they use AI in their workflow.

About the Role

Something is shifting in how software gets built. Stripe ships 1,300+ pull requests per week written by autonomous coding agents, built on infrastructure their engineers created years before LLMs existed. Anthropic reports 70-90% of their code is AI-written. Uber's engineers run multiple background agents in parallel and review the pull requests when they land.

The companies seeing the largest gains aren't the ones with the best models. They're the ones that redesigned their workflows, restructured their teams, and rebuilt their culture around a new division of labour between humans and AI. The gap between we should use AI and we shipped it is where we operate.

You'll work across product teams to figure out where AI fits their workflow, build the right tooling, and get it running.

What you'll do

• Build shared AI tooling. Secure sandbox runtimes, evaluation harnesses, model gateways, prompt management, internal MCP servers. You'll work with OGP's engineering tooling team to extend the developer infrastructure that product teams already depend on, adding the AI layer on top. You'll treat these as internal products: talk to the engineers who use them, write clear docs, iterate on feedback.
• Work alongside product teams. You'll help teams adopt AI-native workflows into how they build and ship. That could mean setting up coding agents and AI-assisted review, or rethinking how a team plans and tracks work. The goal: the team ships their next feature faster because of how they now work.
• Stay ahead of the curve. Between engagements, you'll investigate new models, tools, and techniques. You'll pick what's worth adopting and sell it to the team. The AI tooling landscape moves fast; we need to be tracking it.

What we're looking for

• You've shipped and maintained production software. You've built systems real users depend on, dealt with deployments and incidents, and carried the maintenance that follows a launch. You're fluent across the stack: cloud infrastructure, containers, databases, API design. Working inside other teams live codebases should feel familiar.
• You've built developer-facing tooling. CI/CD pipelines, internal libraries, developer CLIs, APIs, platform services. You've shipped tools other engineers use, whether at work or on side projects. Ideally you've worked on things like model gateways, sandbox runtimes, eval harnesses, or MCP integrations.
• You build on top of LLMs. You already code with Claude Code, Cursor, Copilot, or similar. You've built with LLM APIs — and pushed further into agent architectures, prompt management, tool-use orchestration, or eval pipelines. You have opinions about where autonomous agents work and where they fall apart, formed through use and built-up conviction.
• You learn fast and don't need things spelled out. The AI tooling landscape shifts every few weeks. You'll use different models, different tools, and different patterns six months from now. You can get productive in an unfamiliar codebase in a day, scope a collaboration when the requirements are fuzzy, and know when to ship something rough and when to push for quality.
• You move toward problems without being told. We have no PM, no backlog, no sprint board. We are engineers shaping bets together and deciding how to execute. You need to spot what matters and act on it.

What matters less than you'd expect

• Years of experience. We care about learning velocity and agency. An engineer with a strong portfolio of shipped products will stand out over one with years on a resume.
• Specific frameworks or languages. We work across multitude of teams with different operating stacks and philosophies. Getting productive in any codebase and team quickly matters more than depth in one framework.
• ML or data science background. We're improving how engineers work, not training models. Applied AI fluency beats theoretical ML knowledge.
• Government experience. The role is about AI enablement and engineering, and being a challenge to the status quo.

Working at OGP

Ownership and Autonomy — Beyond technical tasks, this means having autonomy to voice opinions and suggest future directions. Building what you believe in drives the creation of great solutions.

Rapid Prototyping — We prioritise testing ideas over debating them, allowing us to identify problems quickly and convey possibilities to others. This fail-fast approach accelerates innovation.

Flexibility — We provide flexible work arrangements based on what works best for our team and its members, fostering a productive and supportive environment.

Continuous Learning — Drive professional growth by exploring new architectures, frameworks, and technologies. With dedicated yearly learning months and hackathons, continuous learning is essential as we tackle innovative ideas and challenges.

Mission-Driven — Our work is dedicated to the public good, aiming to improve the lives of Singapore citizens. We focus on creating impactful solutions that enhance the community and the place we call home.

Projects OGP has worked on include:

ScamShield — iOS and Android mobile apps, enhanced in August 2024, help users check and report suspicious communications. The apps automatically filter known scams using an AI-powered machine learning classifier. An Admin Dashboard also allows police to verify reported scams.

ParkingSG — A mobile app alternative to parking coupons. It lets users pay, extend, and refund their parking sessions just using their phones.

RedeemSG — Helps the Singapore Government to create, send and track redemptions of digital vouchers easily.

COVID-19 Vaccination — A suite of systems built to enable Singapore's national vaccination campaign for COVID-19. This includes informational sites, appointment booking systems, and records management systems.

data.gov.sg — An open repository of all the Singapore Government's public data. It helps people understand the data using visualisations and articles, and provides real-time APIs for developers to use.

Isomer — Provides government agencies with an easy-to-deploy static website building and hosting service to create usable, secure and faster informational websites quickly.

FormSG — A form builder tool for agencies to self-service and create online forms that capture classified data, with the goal of replacing paper forms.

PaySG — Was developed to enable digital payments for government services, and was initially used for payments for COVID-19 swab tests and Stay-Home Notices for incoming travellers to Singapore.

An overview of other OGP products can be found on our website https://open.gov.sg