How a designer became a top engineer
Katie transitioned from designer to top-performing engineer, ranking in the 94th percentile for code throughput across the entire R&D organization. Her success stemmed from technical curiosity combined with supportive engineer mentors who reviewed her code and helped her improve her craft.
Summary
The transcript discusses Katie's remarkable career transition from designer to highly productive software engineer. Despite lacking formal software engineering training, Katie has become one of the company's top performers, measured by PR throughput into production—surpassing classically trained engineers across the R&D organization at the 94th percentile. When asked about what made her exceptional, Katie attributed her success to two key factors: first, she possessed greater technical curiosity than most designers, giving her a natural inclination toward technical problem-solving. Second, and most critically, she had access to a supportive team of three to four engineers who actively invested in her development by reviewing her code, providing constructive feedback, teaching her how to better leverage AI tools like Claude, and crucially, helping her develop the judgment and taste necessary to distinguish between good and bad code in production environments. This mentorship created a virtuous cycle where she could rapidly improve and contribute at a high level.
Key Insights
- Katie ranks in the 94th percentile for PR throughput into production across the entire R&D organization, outperforming classically trained software engineers despite starting as a designer
- Katie attributes her success to having greater technical curiosity than typical designers, suggesting technical bent is a learnable trait that differentiates performers
- The presence of three to four engineers willing to actively review Katie's code and provide feedback was identified as the most important factor in her development
- Engineers on Katie's team taught her not just how to code, but how to better prompt AI tools like Claude and develop taste for what constitutes good production code
- The speaker indicates Katie's case is an anomaly and poses the question of how to replicate her success across design and product management functions
Topics
Transcript
[0:00] Katie the designer is shipping stuff into production. Her code is the skeleton on which a lot of this functionality is being built. Is she technical? Was she a software engineer? How did she come to shipping code? What blew your mind about her role in [music] this team? >> What blew my mind was that she would turned [music] into this incredible engineer. Across our entire R&D org, she is in the 94th percentile of true throughput, which is a measure of how many PRs you're landing into production. And that includes every single engineer, [0:30] like classically trained software engineer in the company. I asked her, "What is special about you, and how do we get more…
Full transcript available for MurmurCast members
Sign Up to AccessMore from How I AI
Intent engineering beats prompt engineering every time
The speaker emphasizes the importance of integrating AI, specifically Claude, into daily workflows by creating reminders to seek its assistance. This approach allows for collaborative problem-solving, transforming static prompts into dynamic conversations.
Claude Code for normal people: skills, voice mode, and how to collaborate with AI
The episode discusses how AI tools like Claude can enhance productivity and collaboration in business and personal tasks by enabling users to focus on intentions rather than complex prompting. Grace Clark offers insights on her projects and how she built tailored solutions to improve communication and client relations using AI.
Running evals on your internal agents is how you keep them trustworthy over time
The evaluation of internal AI agents is crucial for maintaining their reliability. An internal eval platform allows engineers to review agent performance regularly and ensure their responses are accurate, particularly in critical areas like coding.
You don't have to review every AI-generated PR
The speaker describes an AI agent framework that automatically reviews pull requests by scoring them across six risk factors using markdown-based instructions and skills. PRs are classified as low (0-24 points), medium (25-64), or high risk (65+), with only medium and high risk requiring human approval.
Voice is the highest-bandwidth input channel for AI
Voice interaction with AI addresses the common problem of 'blank chat window syndrome' by enabling users to naturally dump context through speech rather than carefully constructed text prompts. Voice-to-voice AI interaction mirrors the experience of delegating to a human assistant, making it a higher-bandwidth and more natural way to communicate with LLMs.