Instagram story viewer> @michellescomputer> Tagged

Michelle Lawson

@michellescomputer

189.3K

followers

280

following

✨ Turning your dreams into reality
Signed, your fave AI engineer 🤓
💌
[email protected]

POSTS STORIES REELS TAGGED

I’ll let you connect the dots.

@openai DevDay 2026

#openai #ai #sanfrancisco #devday

2k

54

Who would’ve known that going to a film screening alone on October 30th of last year, after flying into LA just for it, would lead to meeting Cat? I introduced myself, and she so kindly offered to sit with me for the movie. Little did I know that one introduction would turn into a lasting connection, and lead to me being part of this moment: the official housewarming party and launch of @catlabs

When I started building @reeldata.app after leaving LA, I had no idea what would follow. It’s insane that this path has brought me back to LA, led me to incredible communities like Cat Labs, and even given me the opportunity to present my work. The people I’ve met along the way have been transformative.

Cat and Maggie, thank you for seeing me, inviting me in, and building such an incredible, supportive community. I’m forever grateful, and so excited to be part of it and to see where it goes from here.

One thing I’ve realized is that a lot of the people doing really cool things are also incredibly humble, kind, and caring. And Cat has a way of bringing exactly those kinds of kind, caring, innovative, creative people into one room.

Congratulations @askcatgpt you’ve built something pretty incredible. Thank you for letting me be a part of it ❤️🙏🏾💥

***it’s Cat not Kat 🫠

24

9

Waitlist in bio ❤️🔗

59

5

Last October I gave 1 person the chance to try cultivated meat

Last week…500 of you

Super grateful to everyone who came out to my first event 🙏 🙏 was incredible meeting all of you and sharing my first YouTube video (releasing soon!)

Still in shock. I started posting content just a year ago, no community, no videos, and now we’re serving 100 pounds of cultivated meat in a night. Insane.

Big things ahead 🐔 🐔

718

22

The projects that get you hired don’t run in the cloud.

Run open-source AI models and AI agents locally on the Razer Blade 16, powered by NVIDIA GeForce RTX 50 Series GPUs. No cloud credits, no data center, no waiting on a queue and just enough GPU to train, test and ship your own projects between classes, while prompts, files and research stay private on your device.

Back to school, built for STEM. Explore the Razer Blade 16 via the link in bio.

351

27

AXIOM is an AI-native engineering intelligence platform,the persistent semantic memory and coordination layer that sits underneath every AI coding agent and every human engineer in a software organization.
AI generates 70–80% of code at forward-leaning engineering organizations. The tools that generate that code — Claude Code, Cursor, GitHub Copilot, and custom agents — are powerful, fast, and capable. But they have zero memory between sessions. Every session starts from scratch. Multiple agents working the same codebase contradict each other, overwrite each other's work, and silently accumulate conflicts. Nobody can answer the question: "Why does this code work the way it does?" There is no audit trail. There is no institutional memory. There is no shared context. AXIOM solves this entirely. It is a platform with seven production-grade components: a living semantic knowledge graph of the entire codebase, a universal agent memory API that every AI tool plugs into, a cryptographic intent audit trail for every AI-generated change, a proactive architecture guardian that detects drift and predicts conflicts, an org intelligence layer that maps expertise and routes reviews, a full-featured dashboard, and a complete enterprise deployment package with RBAC, SSO, and SIEM integration.
#developer #opensource #ai #

7

1

Are you a beginner learning to code? We gotchu 🙂‍↕️🙂‍↕️
We have LOTS of workshops to help you prepare before the hackathon.
Check out @cs.girlies for our workshop marathon schedule ✨

Are you an experienced technical baddie wanting to stack your portfolio? Come join us!! 🤩

Wherever you are in your journey, we’d love to have you at our hackathon!!

Comment “hackathon” and we’ll send you the details 🩷

365

211

We’re so so so excited for this one!!

A HUGE Thank You to @ubeboobey_ for being our July Workshop Guest and sharing her knowledge with us 😄

The @cs.girlies team have taken a sneak peek at the slides and we’re already obsessed 🤭

If you want to learn about creative technology, be inspired to start new projects or simply expand your field of knowledge in tech, comment “Creative” and RSVP!! 💕

We’ll see you thereeee ✨🫶

730

461

Hello! Hello! 👋
Meet the @cs.girlies team 😃😄

For the past two years, we’ve built a community to make tech education accessible ✨for Gen Z, by Gen Z 😉

This August, we’re so so excited to put together our upcoming Tech for Wellness hackathon and provide the space for you to learn, build and create projects with impact! ✨

If you’d like to learn and build, comment “Hackathon” to RSVP for our Hackathon in August!! 💕

667

75

First one to talk about Claude, Codex, AI has to pay...

975

49

The video presentation

We ran into one significant issue during the final day: our Anthropic API credits ran out unexpectedly during the demo recording window. This meant the four Band agents (which use AnthropicAdapter and direct Claude API calls for extraction, risk assessment, and decision-making) could not process new pipeline runs for the video. The governance services themselves continued operating normally. Iron-Thread, AgentID, ChainThread, PolicyThread, and ThreadWatch all stayed live on Render with our keep-alive infrastructure running.

We recorded the video demonstration using the confirmed end-to-end pipeline run already in our Supabase database (chain ID: loan-f01cfb31) and narrated over the live governance console dashboard, which shows all six panels of real data from that run — validation runs, handoff envelopes, compliance evaluations, agent identities, and the signal feed.

The governance dashboard is always live at https://gooverne-agents.lovable.app/ and reads real data. The five governance services are always live on Render. The four agents run locally and require ANTHROPIC_API_KEY to operate — judges who want to run the full pipeline can clone the repo and supply their own key.

Both repositories are public:
Main codebase: https://github.com/eugene001dayne/governed-agents
Dashboard: https://github.com/eugene001dayne/agent-oversight-console

The core product is the governance layer, not the demo workflow. The loan application is the vehicle. Iron-Thread, ChainThread, PolicyThread, AgentID, and ThreadWatch are the product — and they are all live, working, and writing real data to Supabase right now.

#aiagents #developer #hackathon

2

0

Submitting to track 3. Band of agents hackathon
Wish me luck.
"AI agents are being deployed into enterprise workflows right now. Loan reviews. Compliance checks. Contract analysis. But here is the problem, those agents operate with no governance layer. They pass data to each other blindly. Nobody verifies the handoff happened correctly. Nobody checks that the output follows the rules. Nobody confirms which agent is actually sending the data. And there is no audit trail.
We built Governed Agents. The governance infrastructure that sits between agents. Not the agents themselves, the layer that makes agents trustworthy enough to deploy in regulated environments."
#opensource #hackathon #aiagents

11

0