Abstract

In this talk, I will introduce Spec-Driven Development (SDD) in the era of large foundation models and autonomous AI agents, arguing for re-centering the Software Development Life Cycle (SDLC) around specifications to improve alignment, reliability, and long-term maintainability. By placing structured, semi-formal specifications at the core of development, we can ensure higher fidelity between intent, implementation, and system behavior. I will use Kiro (https://kiro.dev/) as a concrete example to demonstrate how development begins with specifications that systematically guide code generation, validation, and evolution. We will explore how SDD supports both greenfield systems, by enforcing clarity and architectural rigor from inception, and brownfield systems, by extracting and leveraging specs from legacy codebases for feature development, bug fix, and modernization. I’ll discuss how coding agents grounded in specs can improve correctness, reduce hallucination, and enforce invariants, while also highlighting open challenges such as spec completeness, ambiguity, trust, and semantic alignment. Finally, we’ll discuss the need for offline benchmarks, high-quality {Code, Spec} datasets, and rigorous evaluation metrics, and conclude with a forward-looking perspective on how specifications may evolve into the primary artifact of software engineering, with code becoming a compiled byproduct of intent.

Speaker Bio

Rajdeep Mukherjee is a Tech Lead and Senior Applied Scientist at Amazon, leading developer tools such as Kiro that power developer experiences both inside and outside Amazon. Prior to working on Kiro, Rajdeep developed AI-based coding assistant products such as Amazon Q Developer Customization, as well as code review products such as Amazon CodeGuru. He earned his PhD in Computer Science from University of Oxford, specializing in formal methods and program analysis for system correctness. Prior to joining Amazon, Rajdeep worked in the Electronic Design Automation industry, where he built formal verification tools at Cadence Design Systems that check the correctness of hardware, software, and safety-critical systems. He has also held research roles at Arm Cambridge, Microsoft Research, IBM Research, and Diffblue. His expertise spans coding agents, LLM applications, static analysis, program synthesis, and formal verification systems.