We're building Aide, an AI-native, privacy-first IDE. With the advent of AI, we believe there is an opportunity and necessity to re-imagine the IDE to be a place where developers and AI are both first-class citizens, and AI can pair-program as well as complete tasks independently.

We're currently a team of two, based in London, United Kingdom. Between us, we have been a tech lead at Meta for testing infrastructure, ACM ICPC world finalist, built payments infrastructure for 150k merchants and Open-Source contributor to critical infrastructure, including gRPC and the Bazel build system. We love dev tooling, and enjoy thinking about the future of programming from a meta level.

We are looking for full-stack developers who believe in the Shift Left paradigm and Ask and Adjust paradigm, ensuring that the most important developer tool for the coming decade is a seamless integration between LLMs and the developer working together in the editor.

Our tech-stack is divided between:

• Sidecar: A powerful rust binary which is the AI brain, humming along on the client side and running locally
• VSCode: The editor built by Microsoft, which we have forked and are changing to enable new experiences

Along with the team you will be responsible for:

• Help to resolve user issues on Discord and other channels
• Build quickly and with speed by choosing the most appropriate of abstractions
• Show not tell, demonstrating what can be possible with an end to end ownership of the tech-stack from the editor to the inference layer of the LLM.

You will feel at home if:

• You really, really care about users. We make sure that each user problem is resolved and anyone how has issues has a quick and efficient resolution.
• You can research your way to new ideas and ship fast and think using first principles
• Build for the current technology but keep an eye on the future which is coming, FAST!
• Think like a philosopher, but code like a hacker :)
• Want to work in person in London!

One of our core principles with the editor, is that it needs to be privacy-first, and be ready for a future where AI models will be running natively on developer machines. Moreover, the editor is a latency critical application because programmers tend to work fast and expect extreme accuracy.

Keep that in mind, our tech stack is currently divided as:

• The editor, currently a fork of VSCode, is an electron app written 100% in TypeScript with very minimal external dependencies.
• The AI sidecar, which is packaged with the editor, is written in Rust.
• We run our custom models on RunPod, and have the update server (written in Python) for the editor deployed to GCP.