Scott Williams

AI @ Tenex

[Arman Hezarkhani @ArmanHezarkhani]

https://t.co/e3wKzSBYnS

[Arman Hezarkhani @ArmanHezarkhani]

Google search is already dead

[Arman Hezarkhani @ArmanHezarkhani]

https://t.co/PvB8pFbYnT

[Arman Hezarkhani @ArmanHezarkhani]

Andrew Ng (@AndrewYNg ) is your favorite ai scientist's favorite ai scientist

https://t.co/NDB2NffHHV

[Arman Hezarkhani @ArmanHezarkhani]

We closed 7 figures in ARR with these logos

I logged in to canva today for something random

Saw my initial designs for tenex's logo

If you're building a business, just go sell

[Alex Lieberman @businessbarista]

After selling Morning Brew for $75m, I never thought I'd build a media company again.

I was wrong. I am.

The media company is Tenex Media. It's the digital destination for AI education & applied AI, and it sits on top of @tenex_labs, the applied AI consultancy & lab my cofounder & I launched 15 months ago.

We're actively hiring for key founding team roles at Tenex Media, which I'll share below.

And in sharing these roles, you'll also get a sense of our strategy—what feels like lessons learned from Morning Brew & what feels like new, unexplored territory.

1) We're hiring 2-3 Full-Stack AI Creators.

Personality-driven content is becoming a bigger and bigger part of consumers’ media diets. The democratization of creation & distribution means that people can attract as much attention as institutions & consumers have always longed for content that feels personal/relatable.

And we're seeing this take shape in the AI space with personalities like @karpathy, @bcherny, @signulll, @levie, @jainarvind and others.

You have maybe the best job in the world as a full-stack AI creator. You build with AI. You talk about what you build. You teach people how to use AI through content. And you train F2000 enterprises on AI.

2) We're hiring a Content Engineer, Editorial.

Brand-driven content isn’t dead. The expectations of it have just changed & it serves a more important role in media than ever before.

There is one foundation, two channels & three franchises that you will own in your role.

The foundation: the content machine. We've developed an internal AI system that allows us to create editorial at scale without sacrificing voice or quality. You will use & improve the content machine to become a one-person wrecking ball.

Two channels: website and newsletter. 30% of our inbound MQLS in Q2 came from AEO. That's crazy because we haven't done jack shit to optimize our website or long-form content. Next, is newsletter. I'm a little biased. But great editorial newsletters have never been more important to own & nurture your audience. You'll take our newsletter, Ultrathink, from 1x to 3x per week and make it the best applied AI newsletter in the world.

Franchises:
- Ultrathink: tenex's editorial newsletter
- Hey chat: site essays answering the most pressing questions enterprise executives have around AI transformation
- LabNotes: site essays sharing step-by-step playbooks to understand AI concepts & how to apply the technology in real business contexts

Content Engineering is to modern marketing what DevOps was to software. A new discipline that makes scale, governance, and speed possible without sacrificing craft.

For more info about the roles, check out the JDs below...

[Dan Zakon @dan_zakon]

Almost every important decision in an AI product now lives in the agent harness.

The harness is the system around the model that controls its instructions, tools, context, memory, permissions, and how it moves through work. Those choices decide how an AI product behaves and how much of the system a developer can inspect and change.

The harness is becoming one of the most competitive layers in AI. OpenAI, Anthropic, Google, and Cursor are all building their own harnesses, each with a complete opinion about how an agent should use tools, manage context, ask for permission, and work with a developer.

They have obvious advantages because they can tune models to their specific harnesses and build polished experiences around the entire loop.

But Pi is making a different bet. It is an open-source harness built around a small, simple core, with a modular extension system that lets developers understand the loop, change it, and embed it inside their own products.

That approach is already producing real results. Pi placed ahead of Claude Code on terminal-bench 2.0 with both running Opus 4.5, and it is becoming the foundation for fast-growing projects like OpenClaw and Flue. Startups are beginning to adopt it as the core of their own agents too.

I think this is a preview of how the harness layer develops. Proprietary harnesses will keep getting more capable, but developers will increasingly want a core they can understand and reshape around their own product.

My bet is that Pi's simplicity and extensibility will win as a design principle for harnesses over time. A small core with clean extension points can absorb new models, tools, and product requirements without forcing every agent into the same fixed system.

[Arman Hezarkhani @ArmanHezarkhani]

https://t.co/0eX4B2YOz6

[Arman Hezarkhani @ArmanHezarkhani]

Tenex is the fastest moving company in the world.

I'll give you an example.

Last week, we made a decision to hire interns.

2 days later, our Director of Ops @BrettFranklin_ had a team of 5 interns working in our office. They even got a full onboarding program.

The time from decision to action. That is speed. For us, it was 2 days.

If you want to work in this kind of environment, reach out. We're hiring a ton at @tenex_labs

[Raghav Dixit @_raghavdixit_]

https://t.co/RVnYCtg2wO

[Alex Lieberman @businessbarista]

Just met with one of our engineering leaders. Some takeaways about AI engineering best practices + more:

1) Spec writing and strong reading comprehension are two of the most valuable skills in ai engineers today.

2) Being hyper structured and opinionated in engineering workflows is how you get probabilistic models to behave deterministically when you want them to and also get models to spend tokens efficiently.

3) Creating a standardized schema/metadata on markdown files in your workflows allows you to make non-software tasks verifiable which allows you to close the agent loop more successfully.

4) One of the bigger behavioral changes in knowledge work is learning to thoughtfully structure/organize your files like good engineers have always done to get the least entropy from models.

5) Building a strong immune system around markdown files is important. As a workflow evolves and gets more reps it’s easy for specs to get bloated with conflicting guidance/unnecessary rules. Hermes agents solves this with a thin memory layer. There’s still a lot of optimization to be done with memory/markdown autophagy.

6) The key ingredients of our engineering process are CLI + deeply opinionated folder/file system + markdown metadata + coding agents + linear as source of truth.

7) Our mental model is always how can we make sure the agent has one way to do things and can validate its output.

[Jake Klaristenfeld @jake_klaris]

The fastest way to make an 85-year-old hang up on your AI? Respond too fast.

We built voice AI for seniors. Everyone in voice optimizes for low latency. Production taught us the opposite lesson.

[Scott Williams @swill1ams]

@GallanisAlex lmaoooo

[Scott Williams @swill1ams]

Def never been that friend

[doomer @uncledoomer]

your friend will be obsessed with vibe coding and youll ask them what theyve vibe coded and itll just be this

[Arman Hezarkhani @ArmanHezarkhani]

Extremely proud of our work with Advantage Solutions and @AnthropicAI.

Advantage is a people-first business in a dynamic market. Dave Peacock (CEO) and Bethany Miles (CAIO) are market leaders in their AI roll out.

Their roadmap shows a balance of ambition and practicality that has driven measurable results. And it's backed by the power of Anthropic's platforms.

We're incredibly excited about the future of these partnerships.

https://t.co/y6QwK4EHGF

[Scott Williams @swill1ams]

@businessbarista ai

[Alex Lieberman @businessbarista]

I’m in a text group with 30 enterprise execs driving ai transformation at their companies.

It is the #1 community I’ve become a part of this year.

Battle tested AI tools, challenges, and wins are shared constantly and we are starting to meet as a group monthly for live discussion.

I want more of the energy and value this group provides, but I don’t want to dilute it, so I'm spinning up a few more of these this quarter.

Each group is capped at 30 people and the rules are simple…

No sales pitches. Just smart people sharing how they’re using AI personally and how they’re transforming their org with AI (successfully and unsuccessfully).

If you want to join one of these groups, reply with "ai,” and if it’s a good fit I’ll DM you an invite.

[Scott Williams @swill1ams]

@BrettFranklin_ dont tell people about the universal context dump

[Scott Williams @swill1ams]

Hot take - Data engineers are some of the least appreciated technologists.

The hard work they do often goes unrecognized until something breaks, and an angry message is sent out.

What people don’t realize it that their work is the backbone of AI (and many other industries). A model is only as good as the data feeding it, and getting clean, correct data to the right place at scale is the part almost nobody sees. The dashboard gets the credit. The pipeline underneath it did the work.

It also takes a rare mix of skills. A good data engineer is part software engineer, part database expert, part distributed systems engineer, part operator, holding correctness, cost, and scale in their head at once.

If you're a cracked Data Engineer, Tenex is hiring.

https://t.co/UyiJzSsqht

[Dan Zakon @dan_zakon]

The modern FDE role is closer to being a founder than being a consultant.

A traditional consultant will diagnose the workflow, a solutions engineer will explain the architecture, and a software engineer will build a feature based on requirements.

But the FDE has to do all of that while the requirements are still moving.

They have to figure out which problem matters, sell the customer on a concrete path, build the first version, notice what breaks, explain the tradeoffs, and decide which parts of the work should become reusable.

That is much closer to founder work than traditional delivery work. It rewards the person who can understand an ambiguous business problem, choose a path, build the first version, explain the tradeoffs, and keep moving while the requirements change constantly.

In the age of AI, one person can now handle the full loop themselves, understanding the workflow, prototyping the system, testing it with users, and turning the result into the next version without waiting for a large team or internal beurocracy.

That's why the FDE is becoming one of the more interesting roles in tech. It is generalist because the work itself crosses product, engineering, sales, strategy, and operations.

If this kind of work sounds exciting, Tenex is hiring: https://t.co/uab5ABVrxB