In this episode of Humans of AI, I sit down with my friend Eric Zhu to unpack what are the frameworks and architectures powering modern AI products today.

From our times building Semantic Kernel, Autogen, and GraphRAG at Microsoft, we offer a retrospective on the evolution of this space and where we think it'll go in 2026 and beyond.

We dive into how AI systems are actually being built today, why early agent frameworks were brittle, and what’s changing as models, tooling, and abstractions improve. Eric shares insights from hands-on experience building AI systems, including lessons around memory, orchestration, open source, and why building has become dramatically faster than even a few years ago.

We also talk about the human side of all this—how builders should think about careers in AI, avoiding burnout, and navigating a fast-moving field without getting overwhelmed.

Whether you’re an AI engineer, founder, researcher, or just curious about where things are headed, this conversation goes deep into the why behind modern AI systems.

⏱️ Chapters

00:00 – Intro

02:02 - Welcoming Eric Zhu! Experiences building open-source systems from PhD to Industry

11:18 – Early agent systems, frameworks, and brittle abstractions

20:18 – Multi-agent frameworks and why it was hard to go beyond a demo

25:47 – From autonomous agents to workflow orchestration - a trough of disillusionment?

30:00 – Breathing new life into agents with Claude Code

34:35 – "Harness" Engineering and what goes into building agent systems?

39:08 – The importance of data and data representations (i.e knowledge graphs, structured data)

45:57 – Lessons from building in Open Source

56:54 – Office Hours: How to not get left behind in AI

Submit your question to the Humans of AI Office Hours and we may cover it in a future episode!

https://forms.gle/iztuQCjLyUe8kXYRA

🎙️ Connect with the hosts:

Alex Chao

📧 Newsletter: https://alexchao.substack.com/

💼 LinkedIn: https://www.linkedin.com/in/alexchao56/

🐦 X: https://x.com/alexchaomander

📱 TikTok: https://www.tiktok.com/@alexchaomander

Eric Zhu

💼 LinkedIn: https://www.linkedin.com/in/ekzhu/

🐦 X: https://x.com/ekzhuAI moves fast.

We keep it human.

After a long break, the Humans of AI Podcast is back for Season 2!

With a new look, new format, new cohosts, and guests, this season will be one you won't want to miss.

This week my cohost Tula Masterman and I recap the saga that was OpenClaw — the open-source agentic project that took the internet by storm, spawned an AI social network where agents allegedly conspired to create their own language, and ended with its solo founder joining OpenAI.

Plus, we cover new research from Anthropic and OpenAI that reveals what actually happens when humans hand the wheel to agents.

We also explore are computer science majors disappearing? Recent data from the University of California seems to suggest so.

And we wrap with Humans of AI Office Hours, where we cover user-submitted or curated questions worth digging deeper into. This week's question: "When AGI?"

0:00 Introduction to Humans of AI Podcast Season 21:46 OpenClaw and Agentic Systems2:30 The Rise of OpenClaw5:50 Transition from Chatbots to Agentic Systems7:42 The Importance of Agent Memory10:59 Agentic Execution and Actions20:42 Societal Implications and Agent Uprising27:46 Harness Engineering and Agent Autonomy44:47 Domains of Agent Deployment49:25 OpenAI's Harness Engineering Blog1:08:09 Decline in Computer Science Majors1:18:42 Humans of AI Office Hours: When AGI?

Submit your question to the Humans of AI Office Hours and we may cover it in a future episode!

https://forms.gle/iztuQCjLyUe8kXYRA

🎙️ Connect with the hosts:Alex Chao

📧 Newsletter: https://alexchao.substack.com/

💼 LinkedIn: https://www.linkedin.com/in/alexchao56/

🐦 X: https://x.com/alexchaomander

📱 TikTok: https://www.tiktok.com/@alexchaomander

Tula Masterman

💼 LinkedIn: https://www.linkedin.com/in/tula-masterman/

🐦 X: https://x.com/tulamasterman

📱 TikTok: https://www.tiktok.com/@tulamasterman

AI moves fast. We keep it human.

Links to stories:

https://techcrunch.com/2026/02/15/openclaw-creator-peter-steinberger-joins-openai/

https://www.anthropic.com/research/measuring-agent-autonomy https://openai.com/index/harness-engineering

/https://developers.openai.com/blog/skills-shell-tips

https://www.nbcbayarea.com/news/local/uc-system-decline-computer-science-enrollment/4032197/

#humansofai #ai #openclaw #agents #harness #claudecode #codex #openai #anthropic #csmajor #college

After a long break, the Humans of AI Podcast is back for Season 2! With a new look, new format, new cohosts, and guests, this season will be one you won't want to miss.

This week my cohost Tula Masterman and I recap the saga that was OpenClaw — the open-source agentic project that took the internet by storm, spawned an AI social network where agents allegedly conspired to create their own language, and ended with its solo founder joining OpenAI.

Plus, we cover new research from Anthropic and OpenAI that reveals what actually happens when humans hand the wheel to agents.

We also explore are computer science majors disappearing? Recent data from the University of California seems to suggest so.

And we wrap with Humans of AI Office Hours, where we cover user-submitted or curated questions worth digging deeper into. This week's question: "When AGI?"

Submit your question to the Humans of AI Office Hours and we may cover it in a future episode! https://forms.gle/iztuQCjLyUe8kXYRA

🎙️ Connect with the hosts: Alex Chao

📧 Newsletter: https://alexchao.substack.com/

💼 LinkedIn: https://www.linkedin.com/in/alexchao56/

🐦 X: https://x.com/alexchaomander

📱 TikTok: https://www.tiktok.com/@alexchaomander

Tula Masterman

💼 LinkedIn: https://www.linkedin.com/in/tula-masterman/

🐦 X: https://x.com/tulamasterman

📱 TikTok: https://www.tiktok.com/@tulamasterman

AI moves fast. We keep it human.

Links to stories:

https://techcrunch.com/2026/02/15/openclaw-creator-peter-steinberger-joins-openai/ https://www.anthropic.com/research/measuring-agent-autonomy https://openai.com/index/harness-engineering/ https://developers.openai.com/blog/skills-shell-tips https://www.nbcbayarea.com/news/local/uc-system-decline-computer-science-enrollment/4032197/

Jeremy Toeman is the founder of AugXLabs, an AI video technology startup, helping creators make compelling videos at a fraction of the time & effort. Jeremy's had a storied career in the media industry leading product and innovation at places like WarnerMedia, CBS Interactive and more and is now directing his energies towards building a new venture in the era of generative AI.

Check out AugXLabs and make your very first Augie! http://www.augxlabs.com/?via=alex

And use promo code AUGIE1MONTH for 1 month free of premium Augie!

If you would like to be a guest or sponsor the podcast, send me a note at [email protected]

And follow me on other platforms so you’ll never miss out on my updates!

💌 Sign up for my free AI newsletter Chaos Theory: https://alexchao.substack.com/subscribe

🐦 Follow me on Twitter https://twitter.com/alexchaomander

📷 And Instagram! https://www.instagram.com/alexchaomander/

🎥 And TikTok! https://www.tiktok.com/@alexchaomander

👥 Connect with me on LinkedIn https://www.linkedin.com/in/alexchao56/

Charles Packer is a current 5th-year PhD student at UC Berkeley and author of MemGPT an open-source project teaching large language models to manage their own memory for unbounded context! Charles is also a core member of the Berkeley Artificial Intelligence Research (BAIR) Lab and Real-Time Intelligent Secure Explainable Systems (RISE) Lab where his work has spanned reinforcement learning and autonomous driving.

Check out MemGPT and join their Discord community! https://memgpt.ai/

And Charles's personal website: https://people.eecs.berkeley.edu/~cpacker/

If you would like to be a guest or sponsor the podcast, send me a note at [email protected]

And follow me on other platforms so you’ll never miss out on my updates!

💌 Sign up for my free AI newsletter Chaos Theory: https://alexchao.substack.com/subscribe

🐦 Follow me on Twitter https://twitter.com/alexchaomander

📷 And Instagram! https://www.instagram.com/alexchaomander/

🎥 And TikTok! https://www.tiktok.com/@alexchaomander

👥 Connect with me on LinkedIn https://www.linkedin.com/in/alexchao56/

Amanda Bickerstaff is the co-founder and CEO of AI for Education, an organization aiming to help teachers and schools unlock their full potential through AI by providing workshops, trainings, tools and resources to educators and policymakers.

Find Amanda on LinkedIn: https://www.linkedin.com/in/amanda-bickerstaff-edu/

If you would like to be a guest or sponsor the podcast, send me a note at [email protected]

And follow me on other platforms so you’ll never miss out on my updates!

💌 Sign up for my free AI newsletter Chaos Theory: https://alexchao.substack.com/subscribe

🐦 Follow me on Twitter https://twitter.com/alexchaomander

📷 And Instagram! https://www.instagram.com/alexchaomander/

🎥 And TikTok! https://www.tiktok.com/@alexchaomander

👥 Connect with me on LinkedIn https://www.linkedin.com/in/alexchao56/

Sudip Chakrabarti is an early-stage investor and partner at Decibel where he focuses on enterprise and infrastructure companies. Before becoming an investor, Sudip was an entrepreneur and an engineer, starting his first company when he was at Georgia Tech, and his second while in business school at Wharton. He's been in venture ever-since with a particular focus on enterprise and open-source companies.

Find Sudip on LinkedIn: https://www.linkedin.com/in/sudipchakrabarti

If you would like to be a guest or sponsor the podcast, send me a note at [email protected]

And follow me on other platforms so you’ll never miss out on my updates!

💌 Sign up for my free AI newsletter Chaos Theory: https://alexchao.substack.com/subscribe

🐦 Follow me on Twitter https://twitter.com/alexchaomander

📷 And Instagram! https://www.instagram.com/alexchaomander/

🎥 And TikTok! https://www.tiktok.com/@alexchaomander

👥 Connect with me on LinkedIn https://www.linkedin.com/in/alexchao56/

Josh Fechter is the CEO and founder of Squibler, a startup seeking to be your AI Copywriter and Storyteller. Josh is a growth hacking expert, a builder of communities and has written several books on software tech writing and copywriting for personal brands.

Find Josh on LinkedIn: https://www.linkedin.com/in/joshua-fechter/

And check out Squibler: https://www.squibler.io/

If you would like to be a guest or sponsor the podcast, send me a note at [email protected]

And follow me on other platforms so you’ll never miss out on my updates!

💌 Sign up for my free AI newsletter Chaos Theory: https://alexchao.substack.com/subscribe

🐦 Follow me on Twitter https://twitter.com/alexchaomander

📷 And Instagram! https://www.instagram.com/alexchaomander/

🎥 And TikTok! https://www.tiktok.com/@alexchaomander

👥 Connect with me on LinkedIn https://www.linkedin.com/in/alexchao56/

Aayush Mudgal is a Senior Machine Learning Engineer at Pinterest, currently leading the efforts around Privacy Aware Conversion Modelling.

Ayush has expertise in large-scale recommendation systems, personalization, and ads marketplaces. and in the past has conducted research on intelligent tutoring systems, developing data-driven feedback to aid students in learning computer programming.

Find Aayush on LinkedIn: https://www.linkedin.com/in/aayushmudgal/

And for the books that Aayush recommends:

Why We Sleep: Unlocking the Power of Sleep and Dreams (https://amzn.to/48cVdmn)

How to Fail at Almost Everything and Still Win Big (https://amzn.to/3P5AUOG)

An Elegant Puzzle: Systems of Engineering Management (https://amzn.to/3Z91kUA)

If you would like to be a guest or sponsor the podcast, send me a note at [email protected]

And follow me on other platforms so you’ll never miss out on my updates!

💌 Sign up for my free AI newsletter Chaos Theory: https://alexchao.substack.com/subscribe

🐦 Follow me on Twitter https://twitter.com/alexchaomander

🧵 And Threads! https://www.threads.net/@alexchaomander

👥 Connect with me on LinkedIn https://www.linkedin.com/in/alexchao56/

Adam Steinle is the cofounder and CEO of hona.ai. Adam started out as a biomedical engineer, and after stints at Goldman Sachs and Facebook, he returned to his roots to help medical systems communicate so that providers can learn about their patients more effectively.

Find Adam on LinkedIn: https://www.linkedin.com/in/adamsteinle/

And on YouTube: https://www.youtube.com/@adam_steinle

And to read more about Hona AI check out their website: https://www.hona.ai/

For the YouTube channel that Adam recommends: https://www.youtube.com/@jayhoovy

If you would like to be a guest or sponsor the podcast, send me a note at [email protected]

And follow me on other platforms so you’ll never miss out on my updates!

💌 Sign up for my free AI newsletter Chaos Theory: https://alexchao.substack.com/subscribe

🐦 Follow me on Twitter https://twitter.com/alexchaomander

🧵 And Threads! https://www.threads.net/@alexchaomander

👥 Connect with me on LinkedIn https://www.linkedin.com/in/alexchao56/