AI-Native Software Engineer

About Qandaba

Qandaba is an early-stage technology company. We build software products and we help other companies adopt AI-native engineering practices. We are small and fast-moving, and we treat large language model coding tools as a core part of how we deliver software, not as a side experiment. You would join as one of our first engineering hires, working directly with the founder.

TL;DR

Contract-to-Hire · Remote, Philippines · Qandaba

Role: AI-Native Software Engineer

Engagement: Full-time contract-to-hire (6 months, convertible to a longer-term role)

Location: Remote, Philippines-based

Level: At least 3 to 5 years building software, AI-native throughout

Works with: Directly with the founder, as one of the first engineering hires

The Role

We are looking for an engineer who is fluent at delivering software with LLM coding tools such as Claude and Codex, and whose real strength is defining what to build and confirming it was built correctly, rather than typing most of the code by hand.

Above all, we hire for demonstrated judgment and the right mindset, paired with a genuine affinity for working through AI tools. That matters more to us than a long resume or any particular tech stack.

A note before you apply: This role is built for someone who already works almost entirely through LLMs and AI coding and verification agents. If those tools are not how you do at least 80% of your engineering work today, this is not the right role for you. No judgment, it is simply a different kind of job, and applying will not change the fit.

This is a different job from a traditional software engineering role. The coding tools write most of the code. Your job is to lead them well: turn a rough idea into a clear, testable specification, direct the tools through the build, and verify rigorously that the result does what it should. What matters here is how you break a problem down and your discipline around verification, not how quickly you can write a function from scratch.

A systems integration background fits this work well. Most of what we build is about connecting things: applications to backends, software to hardware, one service to another. You should understand how mobile apps and cloud or backend services work, and how they fit together. You do not need deep specialist expertise in either. You need to be the person who can reason across the whole system and catch where the seams will fail.

One principle runs through all of it: observability. Because the tools, not you, write most of the code, the work is only trustworthy if every change is traceable. That means a clear line from each requirement, through the decisions made and how they were tested, to the verified product, with the points where human judgment was applied recorded along the way. Treating that as non-negotiable for every change an LLM makes is the core of this job.

How You Will Work

• Requirements first. Before code is written, you turn an idea or request into a clear specification: what it should do, what it should not do, and how we will know it works.
• Direct the tools. You drive the LLM coding tools through the build, giving them well-scoped tasks and the context they need, and stepping in with judgment when they drift.
• Verify everything. You design and run the checks: test cases, edge cases, and acceptance criteria. Automated tests are not enough on their own. You apply human-in-the-loop judgment and treat generated code as wrong until it is proven right.
• Make the work observable. Every change carries a record: what was decided and why, how it was tested, and where you applied human judgment. Anyone should be able to trace a clean line from a requirement to the verified result.
• Run the checking agents. You put agents to work on security, compliance, and performance checks for the code the tools produce. You do not need to be an expert in any of those areas. You need the mindset that every LLM-made change earns that scrutiny, and the ownership to make sure it happens.
• Integrate and ship. You connect the parts into a working whole and make sure the seams hold under real conditions.

What You Will Work On

You will move across several products and internal systems, so comfort with variety matters more than depth in any one domain. The work spans:

• Consumer mobile applications and the backend services behind them
• Integration between software and physical hardware
• Internal tools and operating systems that the company runs on
• Reusable components and templates that speed up future builds
• Data and analytics products, including workforce-focused analytics
• AI-assisted content and learning products, including content generation

We are not detailing each project here. The point is the shape of the work: a mix of customer-facing apps, internal tooling, and newer AI-native products, all delivered with LLM coding tools and held to a high verification bar.

What We Are Looking For

• Heavy, hands-on use of LLM coding tools (Claude, Codex, or equivalent) in real software work. You can talk concretely about how you use them, where they fail, and how you compensate.
• A requirements and verification mindset. You are strong at scoping work, writing clear specifications, defining acceptance criteria, and testing rigorously.
• An observability and traceability instinct. You believe every change an LLM makes must be traceable from the original requirement to the verified result, with the decisions, tests, and human judgment recorded along the way. You are comfortable putting agents to work on security, compliance, and performance checks, and you treat that scrutiny as non-negotiable, even though you are not expected to be an expert in each area.
• A systems integration background. You have connected systems, services, and components, and you think naturally about interfaces, contracts, and failure points.
• Working knowledge of mobile and backend. You understand how mobile applications and cloud or backend services are built and operate. Deep specialist expertise in either is not required.
• Solid engineering fundamentals. Version control discipline, the ability to read and reason about code, debugging, and basic API design.
• Roughly 3 to 5 years building software, AI-native throughout. We are not looking for a 15-year veteran. We want someone who has been shipping real work and doing it through LLMs, heavily since ChatGPT arrived and more so since Claude Code and Codex.
• Based in the Philippines with strong written and spoken English, and comfortable collaborating on a remote team.

Nice to Have

• Experience shipping mobile apps (iOS or Android) or their backend services
• Cloud platform experience (AWS, GCP, or similar)
• Familiarity with hardware, IoT, or device integration
• Python and/or TypeScript fluency
• Experience defining test strategies or quality gates
• Exposure to building internal tooling or business systems

Engagement and Setup

• Full-time contract-to-hire. The first six months are a full-time contract engagement. If it works well for both sides, it converts to a longer-term role.
• Remote, Philippines-based. You will need a reliable internet connection and a professional working setup.
• Schedule. Some overlap with US working hours for collaboration. The exact arrangement is open to discussion.
• Close collaboration. You will work directly with the founder, with real autonomy and a fast feedback loop.

Compensation

Competitive for the role and the experience level described above. We discuss the specific rate directly with shortlisted candidates.

How to Apply

The strongest signal is evidence that you already work this way. If you have shipped real software using LLM coding tools, lead with that: what you built, how you directed the tools, and how you verified the result. Examples of clear specifications you have written, or test strategies you have designed, count just as much as code samples. Tell us about a time the generated code was wrong and how you caught it.