MisterBiggs

6 ms·

1.
by MisterBiggs 3mo ago
Yup! I feel pretty strongly that every little nit pick and instruction you pass into your model is murdering your output. Having a hook that executes on tool calls is significantly better than telling your agent to follow your repos specifi
2.
by MisterBiggs 3mo ago
Skills for repitition are totally valid. Having a version control skill that explains that I use gitea works great. My point is that asking for a skill that tells us if our program will get stuck before taking on a halting problem won'
3.
by MisterBiggs 3mo ago
This is a pretty neat, I suspect that eventually every skill will have some sort of validation/verification loop like this
4.
You're probably using Agent Skills wrong(notes.ansonbiggs.com)
74 points by MisterBiggs 3mo ago|25 comments
5.
by MisterBiggs 5mo ago
What happens once an agent can reliably get 100% on swebench?
6.
by MisterBiggs 7mo ago
For writing with an LLM you really have to be precise and honest with your AI usage and I don't think the disclosure here does a good job at that. If you sample a few random pages there are huge shifts in style and tone. The future whe
7.
by MisterBiggs 7mo ago
It has nothing to do with running models locally, its perfect because its incredibly cheap, capable, small, and quiet.
8.
by MisterBiggs 8mo ago
I was really hoping for a new form factor or new killer features. Its too bad that the general public can't behave themselves with simple tech like this
9.
by MisterBiggs 8mo ago
I feel like I made a mistake going with subdomains for each project instead of root folders, but here is my landing page https://ansonbiggs.com/
10.
by MisterBiggs 8mo ago
Cool to see that Anthropic and OpenAI are really diverging in their target customer. Anthropic seems much more focused on making tools for pros while OpenAI is aiming at consumers.
11.
by MisterBiggs 10mo ago
Shield AI | Autonomy Applications Manager | Full Time | ONSITE Washington DC Shield AI's mission is to protect service members and civilians by creating intelligent systems. My team needs a manager, its a very well positioned team at a
12.
by MisterBiggs 11mo ago
Shield AI | Autonomy Applications Manager | Full Time | ONSITE Washington DC Shield AI's mission is to protect service members and civilians by creating intelligent systems. My team needs a manager, its a very well positioned team at a
13.
by MisterBiggs 1y ago
Shield AI | Autonomy Applications Manager | Full Time | ONSITE Washington DC Shield AI's mission is to protect service members and civilians by creating intelligent systems. My team needs a manager, its a very well positioned team at a
14.
by MisterBiggs 1y ago
I was hoping for more details about their implementation, I saw ollama as the open source // platform agnostic tool but I worry their recent posturing is going against that
15.
by MisterBiggs 1y ago
Great way to sell some of those empty GPU cycles to consumers
16.
by MisterBiggs 1y ago
I was laid off this year and also saw a massive uptick in offers in July. Based in Colorado, but was looking nation wide and I got more interest out of SF than the rest of the country combined.
17.
by MisterBiggs 1y ago
I think more worrying is devices mesh networking with devices owned by your neighbors.
18.
by MisterBiggs 1y ago
Same experience. I hopped on the beta because I thought the current version was going to be really bad and I wanted to watch them move towards something more functional. Its definitely not perfect but the way that the UI reacts in real time
19.
by MisterBiggs 1y ago
Location: Denver, CO Remote: Open Willing to relocate: Yes Technologies: Python, C++, Matlab, Rust Résumé/CV: https://ansonbiggs.com/resume Email: anson@ansonbiggs.com I'm Anson, an Aerospace Simu
20.
by MisterBiggs 1y ago
Location: Denver, CO Remote: Hybrid Preferred Willing to relocate: Looking to relocate to another large city Technologies: Python, C++, Matlab, Gitlab CI/CD, Docker Résumé/CV: https://ansonbiggs.com/res
21.
by MisterBiggs 1y ago
Lots of comments on the price being too high, what are the odds this is a subsidized bare metal cost?
22.
by MisterBiggs 1y ago
Is this new? I've been using icloud with a custom domain for about a year and just had my first failure today with an address that I've actively been talking to all week.
23.
by MisterBiggs 2y ago
This is really cool, does anyone know how it compares to NASA Trick? https://github.com/nasa/trick
24.
by MisterBiggs 2y ago
There is a lot of value in the development story for me. Over time they've almost entirely stopped talking about progress outside of major milestones which is disappointing but theres only so many years in a row you can say server mesh
25.
by MisterBiggs 2y ago
> average donation at around 150 dollars. That seems really high. Check out the store, its absurd what they are able to charge for completely digital assets that don't even exist in game yet
26.
by MisterBiggs 2y ago
I funded this back on kickstarter when I was still in high school. I remember the excitement that came with every $1mil milestone, it really felt like an incredible movement that was going to produce something really special. I haven't
27.
by MisterBiggs 2y ago
I think the real issue is Google is able to use Chrome to push web standards in any direction they want.
28.
by MisterBiggs 2y ago
The incentive to sell something completely undermines the ability to give you the best information. Floating sponsored links to the top is one thing but telling an AI this is the link the user should click is going to ruin the whole thing.
29.
by MisterBiggs 2y ago
I'm a huge fan of the games he has been a part of so I was familiar enough with his antics to know that Curiosity would end up being nothing, but he really outdid himself here, Peak Peter Molyneux behavior.
30.
by MisterBiggs 2y ago
Not sure why Steam is in the headline as the cause when this is really on Sony. Adding region restrictions well after release was a terrible decision on their part and I think Steam being very lenient with refunds was the best they could do

More ›