The Agentic Review

In this episode of The Agentic Review podcast, Wei-Wei Wu, Co-founder and CEO of Momentic, joins Itamar Friedman and Nnenna Ndukwe to talk about AI testing, independent verifiers, and the future of QA. AI now generates code faster than humans can check it. Wei-Wei explains why the same agent shouldn’t write both the code and its tests, how AI is finally forcing teams to shift left, and why tests are becoming an implementation detail behind plain-English product specs. He also argues that change failure rate hides what customers actually experience, and that quality and taste are the real moat when code costs almost nothing to produce.

In this episode of The Agentic Review podcast, Zichuan Xiong, Global Head of AIOps at Thoughtworks, joins Itamar Friedman and Nnenna Ndukwe to look at what AI is really doing to the cost of running software. Over two years, Zichuan’s team grew from a single RAG use case to 25 to 35 operational agents for enterprise clients. He has seen both the job-level gains and the system-level bill: one customer raised AI adoption 1.5x in eight months while operations costs rose 10x. He makes the case for an operations-first SDLC and explains why semantic drift slips past LLM tracing. Zichuan also argues that turning tribal knowledge into agent-ready context needs a dedicated new role.

In this episode of The Agentic Review podcast, Liz Fong-Jones, Technical Fellow at honeycomb.io and Co-Author of Observability Engineering, joins Itamar Friedman and Nnenna Ndukwe for an open conversation about what happens when your team’s pull request volume more than doubles thanks to AI. Drawing on her two-plus decades in site reliability engineering, Liz shares how Honeycomb went from roughly 30 to 70 PRs on a peak weekday and the honest tradeoffs that came with it.

Are we shipping better software, or just more of it? Emilie Schario, VP of Engineering at Anaconda and Co-founder of Kilo (acquired by Anaconda in July after growing from 0 to 3M+ developers in 16 months), joins Itamar Friedman and Nnenna Ndukwe on The Agentic Review podcast to discuss what actually has to change inside engineering organizations once AI tools show up.

Nirmal Jingar discusses deterministic governance and why wrapping AI in guardrails is essential for production. We explore the need to modernize legacy systems before adding AI, the importance of AI literacy at the board level, and why organizational tribal knowledge remains a critical gap for LLM context.

Building a reliable software factory requires more than just generating code. This conversation focuses on the importance of quality assurance and the evolution of developer experience in the era of autonomous agents. Learn how to implement rigorous entry and exit gates to ensure that AI productivity does not come at the cost of technical debt.

This episode explores rapid enterprise AI adoption and why visibility into “shadow AI” and runtime control are essential to a robust AI control plane. Learn how organizations can securely steer thousands of agents while maintaining the low latency and high accuracy required in production environments.

Software engineering pioneer Paul Duvall joins the show to discuss why traditional pull request processes are breaking. This episode delves into quality checks and the development of deterministic guardrails to manage the surge of AI-generated code in enterprise environments.

Angie Jones on agentic tools, repo-level guardrails, and building software you can trust. This discussion explores the necessity of tailored tool stacks and why human judgment remains the non-negotiable anchor for scaling quality. Learn how junior developers, robust testing, and team-wide standards integrate into an accelerated AI-driven engineering workflow.

AI agents in production and code quality governance are critical for software engineering leaders. Learn context engineering and 12 factor agents with Dexter Horthy, CEO of HumanLayer. This episode explores hard lessons from the trenches of AI-assisted development and why the industry is swinging back to manual code oversight to save failing architectures.