Description
Summary
Do you want to build the layer that lets every app on the device do its best work for the user through Siri? App Intelligence is the on-device system that turns app capabilities into things a user can simply ask for — and makes the assistant reliable enough to be trusted with them. We are engineers, scientists, and problem solvers bringing smarter, faster, and more natural interaction with apps to Siri. We are looking for seasoned AI Software Engineers to join our team and be part of this mission.
Description
In this role, you will be at the forefront of developing software on device that enables Siri to deliver on app Intelligence throughout the OS. We bring app intelligence to Siri by working directly with app teams to understand what their apps can do, modeling those capabilities so an assistant can use them well, and building the on-device software that makes the resulting experiences reliable and fast. We partner closely with the team that builds Siri's planning and execution engine, and we work the app side of that partnership: what apps can offer, how they express it, and whether the resulting experience is good enough to trust.
This is early work. How apps best expose themselves to an assistant is an open question — you will help answer it by designing the interfaces, proving them out with real app partners, and iterating as we learn. Expect real ambiguity, and expect to form your own view of the right architecture and defend it with working software and measured results.
Key Responsibilities
Lead the design and development of software features that enable end to end app experiences in Siri
Work directly with app teams to identify where intelligence makes their app better, and bring those capabilities to Siri
Help define the shape of this layer and bring app partners along as it takes shape
Implement code that is well designed, easy to debug, and highly performant
Diagnose failures in end-to-end app experiences, and work with the Siri team to turn them into measurable quality improvements
Contribute to the shared evaluation and telemetry loop that proves whether a change improved user outcomes
Identify and propose key architecture changes to help features scale
Proactively look for areas of code to further optimize for performance, and hold on-device latency, memory, and power budgets
Leverage agentic coding to enhance engineering productivity, code quality, and efficiency at scale
Minimum Qualifications
4+ years experience developing and shipping production software
Strong proficiency in Swift, Objective-C or C++
Experience building product features on top of large language models — designing tool and function-calling interfaces, constructing context, and validating model output
Fluency with agentic coding tools for accelerating engineering productivity, with good judgment about where to apply them and how to verify what they produce
Exceptional problem-solving skills, and the ability to drive alignment across teams on shared API surfaces
Proven ability to excel in a fast-paced development team
Preferred Qualifications
Experience designing and developing systems that are efficient in resource constrained environments — latency, memory, and power budgets on device
Experience building the scaffolding around models — tool interfaces, retrieval and context construction, output validation — with strong intuition for model behavior and failure modes
Experience building key software components and APIs used cross functionally across different applications and platforms, including versioning and compatibility of interfaces that shipped clients depend on
Experience with evaluation, benchmarking, and telemetry for systems whose output is probabilistic rather than deterministic
Apple