Avanceé

A micro-consulting platform delivering capability architecture & strategic guidance to executives and small/mid-sized businesses — by @arjwright

[Avanceé @AvanceeAgency]

Tweaked this… now better articulated as:

Avanceé is a micro-consulting platform delivering strategic guidance, tailored to the competency maturity of a user or team.

[Avanceé @AvanceeAgency]

What is Avanceé?

A sounding board to those who wish to connect strategy & leadership with novel/innovative technologies, so that they can prosper, or be more resilient, in the face of change

[Dr. Timothy C. Summers @HowHackersThink]

The map got stale. The system didn't.

Full essay, free to read:
https://t.co/8peqMZN4IG

[Avanceé @AvanceeAgency]

Current 💭

What’s just a bit too far out there, but still pretty neat and doesn’t just play “demo” but lands as a “here’s a present thing which genuinely helps a humane-focused person express a door to open the capability of others.”

[Avanceé @AvanceeAgency]

…am liking a direction my recent work has been going: interpretive anchors around concepts like: invitation-based systems, contextual computing, trust surfaces, and human-centered operations

From looking for “work” to “making the work speak.” That shape of things is better, now

[Carsten Munk @stskeeps]

The world is shifting from a scarcity of capability to a scarcity of judgment

[Brilliant Labs @brilliantlabsAR]

https://t.co/qNMtvelwqm

[Avanceé @AvanceeAgency]

They could be dead…

They could also be reframed differently. RT was an informal thing made into a framework. It likely has adaptability by those creative enough to tap it

[Robert Scoble @Scobleizer]

Reshares (formerly known as retweets) are dead.

An end of an era in social media.

[BuccoCapital Bloke @buccocapital]

This is why everyone is so exhausted at work

Your company is scrambling to adopt AI in order to keep up with its competitors. You are scrambling to adopt AI to maintain parity with your peers.

Nobody is gaining an advantage, everyone is just getting fitter

[Avanceé @AvanceeAgency]

One of our partners is adamant that fundraising should be at the top of political priorities…

…in learning this space, we find that funding is more of a ripple of trust and accountability. Neither of these is easily grasped but is easily manipulated

[Brilliant Labs @brilliantlabsAR]

https://t.co/P0aLki8c8e

[Avanceé @AvanceeAgency]

😏

…yea, seems like it

[Shashi (シャシ) @shashpicious_]

we’re not really “designing” right now,
we’re just constantly switching contexts trying to not get left behind

every week there’s a new tool claiming to be the future → paper, pencil, magicpatterns, magicpath… now noon shows up with $44M and changes the narrative again

so instead of going deep, everyone’s just sampling everything

trying prompts here, generating screens there, tweaking in figma, jumping to code, back to AI again

half the industry is already inside code editors
the other half is still figuring out which tool is even worth committing to

fomo is doing more damage than we realise

because depth needs stability
and right now the stack itself is unstable

so no one is mastering anything
everyone is just trying to be early

eventually this will settle and a default will emerge

till then, we’re all just beta testers pretending to have a workflow 👀

[Avanceé @AvanceeAgency]

feeling good after the work which went into the concierge-type app talked about earlier this week…

https://t.co/42HEqBMV3w

[Will Burns 🍥 @AeonixAeon]

Imagine you’ve bought a brand new Lamborghini, only to find that you’ve been driving in 1st & 2nd gear with the parking brake on the entire time. That’s where most #AI interactions are today.

But we can fix this.

https://t.co/EmqaIJ2E6H

[Avanceé @AvanceeAgency]

When my @MuseAppHQ corpus can be utilized by an LLM, similar will happen

[Andrej Karpathy @karpathy]

LLM Knowledge Bases

Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So:

Data ingest:
I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them.

IDE:
I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides).

Q&A:
Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale.

Output:
Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base.

Linting:
I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into.

Extra tools:
I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries.

Further explorations:
As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows.

TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

[Avanceé @AvanceeAgency]

🧐💭

[Jim Prosser @jimprosser]

https://t.co/k4zNnlABHU

[Avanceé @AvanceeAgency]

Looking forward to a bit of reduced friction with tomorrow’s “Avanceé Reads” post due to a few tweaks. Will know if things work once it goes live kinda thing

Looking back, been a week of infra-type stuff that’s for sure

[Avanceé @AvanceeAgency]

This mirrors our use of @MuseAppHQ

[Henrik Karlsson @phokarlsson]

In Ancient Greece, the intellectual class had a practice called hypomnēmata (ὑπομνήματα), which was a notetaking system where they kept quotes from books they had read. It was a tool for meditation. Every day, ideally, they would open the hypomnēmata and look for a passage relevant to something they were struggling with, and then they would meditate on that—unpacking it, making the idea top of mind, ensuring it was alive in them. If they needed courage, they’d meditate on a story that made it real for them what it meant to act with courage, and so on.
The idea was that over time, the important insights they gathered by reading would be transformed into character, a habit, something deeply ingrained into their way of thinking and acting. It is hard to keep these complex and anti-memetic ideas alive in us, but certain pieces of writing can make us wake up, and if we return to them and meditate on them, they can become part of us.

[Avanceé @AvanceeAgency]

Each time this comes across the timeline, I think of @MuseAppHQ and how folks who use it might be ahead of things, w/o some of the plumbing

[Arvind Jain @jainarvind]

https://t.co/Y7PmieSGS7

[Brilliant Labs @brilliantlabsAR]

In a new device category, shaping the entire stack — atoms, electrons, bytes, and photons — is the only way to enduringly innovate.

This process is as painful as it is creatively rewarding — over the course of developing a new device, we build and rebuild the ENTIRE stack multiple times.

Because we believe innovation is more than the object — it is the whole system, an articulation of a ‘way of building’: your purpose, culture and values, supply chain, business model, and yes, product.

🚀It is full stack work and requires relentless systems thinking — it is struggle, it is emergent, and it is built off hard-won learning… not viral stunts.

[Eric Jiang @veggie_eric]

Every company should hire an internal AI transformation person. No need for a fancy title like Head of AI. Just give them full latitude to clean up inefficiencies across sales, hr, finance, etc. There's so many manual workflows and arcane bs that can easily be fixed with LLMs

[Avanceé @AvanceeAgency]

A few recent posts:

https://t.co/0lX2SySjAd

https://t.co/gAgVNnuqHA

Quiet ruminations to start the year