Leading AI development transformation
Yonah Wolf
I'm building, learning, and leading the shift to AI-native engineering — mentoring developers, writing code alongside them, and bringing AI into the products we ship.
I'm building, learning, and leading the shift to AI-native engineering — mentoring developers, writing code alongside them, and bringing AI into the products we ship.
Hi, I'm Yonah.
I have been building great software and great teams for the past 2 decades. I have written software for one of America's largest retail chains, a hedge fund with billions under management, one of the world's largest non-profits, and two of the largest 3PLs not named Amazon. I have also built and managed great teams at every step of the way.
Professionally, I operate on two guiding principles — you can read them on the right. I believe the best software comes from teams that feel respected, empowered, and challenged. My job is to remove obstacles, set clear direction, and create the conditions for greatness.
Today that job includes leading my teams through the AI transformation of how we build software — and getting my own hands back in the code to do it.
Outside of work I'm a dad, a husband, a youth coach, and a Judo instructor. You can read more about that — plus my bookshelf and writing — off the clock.
"Don't be mean to your users."
Try to truly understand your users' pain, and the problem that your software solves for them. Projects always take longer, features shift and change — but at the end of the day, the value of your software is how it helps its users get their jobs done.
"Take care of your people and they will take care of you."
Treat your team with respect and give them the tools, training, and environment to do their best work. Go to bat for them when warranted. And don't shy away from hard conversations — giving honest, constructive feedback is how people grow.
AI is changing how software gets built, and I'm leading that change from the inside — reshaping how my teams plan, code, test, and ship, contributing to our repos myself, and putting AI to work in the products our customers use.
Over the past 3.5 years across my two most recent roles, I've led multiple engineering teams — up to 12 developers — coordinating delivery across concurrent projects while managing vendor and offshore consultant partnerships. Here's how I operate as an engineering manager.
Managed multiple concurrent teams of up to 12 engineers — setting direction, balancing workloads across squads, and creating the conditions for people to do their best work.
Planned and coordinated delivery across parallel projects and stakeholders — keeping scope, timelines, and priorities aligned from kickoff through ship.
Built and managed relationships with external vendors and offshore consultants — integrating distributed talent into cohesive teams and holding partners to a high bar.
Invested in engineers through honest feedback, coaching, and hands-on mentorship — now including how to work effectively with AI — because taking care of people is how teams compound over time.
Many of my work projects remain the property of my respective employers. Click any project to read more.
AI-powered document analysis for Cybersecurity Compliance
OpenAI-based parsing of hazardous materials data
Regional demand forecast models for customers
Algorithm for optimized warehouse routing
Slack notifications for warehouse management
Simulations for Brexit and the 2016 elections
Real-time financial risk metrics and alerts
Software I build on my own time — to stay hands-on and try new tools.
An iPhone and Apple Watch app that schedules runs into open gaps in your calendar, adjusts for weather and meeting changes, and syncs with Apple Health and Strava.
A sign-up and scheduling app for my synagogue's minyan — filling services, chazzanim, Torah readers, and aliyot.
An AI-powered tool that generates curated reading lists from podcast guest appearances — with library availability checks and Amazon links.
Rolling out AI across an engineering org? Experimenting with Skills, spec-driven development, or token budgets? I'd love to compare notes on what's working, what isn't, and what's next.