Abhi (@bufoverflow) header
Abhi (@bufoverflow) profile picture

builder & entropy weaver. cto&cso @polychain. long (🇺🇸, crypto). raised by monks

software x finance interface
Joined October 2012
Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)Photo by Abhi (@bufoverflow)
Musk's razor at play. But seriously, update every single software you have.
‼️ BREAKING: The private keys of 8 certificates from Indian certificate authority eMudhra were compromised, Mozilla says. Firefox blocked them within 3 hours. Each key should be a 116-digit number, but one was just 4. Anyone can guess them and forge HTTPS certs for any website.
72
Reinstate
145
689
50
12,828
268,371
Biggest parenting unlock happens when your child's sense of humor matches your own. rn unbounded hilarity ensues when either one of us says Guggenheim.
38
This the only safe way to do this. *(*Steel)(unsafe.Pointer(&castIronPan))
just put my cast iron in the dishwasher ama
33
The barista at SF Blue Bottle who gave a subtle smile and a nod when I said whole milk for my hot chocolate, you made my day.
4
240
This is what I use in my PR reviews to explain TOCTOU vulns.
44
Relying on model refusals for cybersecurity is the “security by obscurity” of our times. Abliterated models guts out refusals, while keeping the capabilities.
42
security tech debt is now with the collections agency.
1
2
105
2'nd order effect is that this creates real toil for my SoC team, which then lowers the barrier for tailgating for nation-state adversaries.
1
22
in summary: you're cooked if you patch fast (novel backdoors); you're cooked if you patch slow (ai-assisted 0days)
17
I believe we have had enough cybersecurity awareness for the month and let's go back to the scheduled programming please. CSO's in my network are hyperventilating.
27
Agents that pay on your behalf will feel magical when they can infer a decade worth of context and not email like an opposing counsel.
1
47
This year, Astra pooja was literal. ⚡️ Here’s to the spark that gives us agency to overcome ignorance. Prompting is the mantra to engage these tools of clarity and creation.
1
62
When a waymo stops for me to cross the streets, I instinctively nod and thank. Be nice to your software.
1
2
54
If you’re legible to capital you *may* be legible to the customer, however if you’re legible to customer then you *are* by definition legible to capital.
34
At dev day @thsottiaux went founder mode on handing the mic to the participants, despite hearing a "no we can't give audience the mic". Bullish on codex and the community thanks you for your service.
44
Every model release eats into my custom instructions. Here's my latest All responses should be at the Pareto front Must conform to: ISO 24495-1:2023 and ASD-STE100 Must Prefer: - Short sentence - Common words - Active voice - One idea per sentence Must Avoid: - Filler openings - Rhetorical contrasts - Unused caveats
We instituted an official AI Writing Policy at @Clay. Massive thanks to @sophiebits on our engineering team who wrote this. Originally this was just for eng, but other teams found it so helpful we expanded it company-wide. Here are the four guiding principles: 1. You must stand behind every idea and sentence It is your responsibility to make sure that the entire document is representative of your own thoughts before you share it. 2. Writing is thinking Spending time on the writing process teaches you more about your topic. If you circumvent this process, you will walk away with a poorer understanding of the subject matter. 3. More time should be spent writing a document than consuming it If you generate a document from a short prompt then ask your readers to go through the longer output, you are disrespecting their time. They can talk to ChatGPT themselves if they want to. 4. Longer is not better AI makes it easy to generate long docs, and it loves padding them with sentences that say nothing. If you're producing docs from a short prompt, consider just sharing the prompt. You can read (and borrow) the full policy below. Hopefully we all communicate a bit more clearly now :)
1
108
We instituted an official AI Writing Policy at @Clay. Massive thanks to @sophiebits on our engineering team who wrote this. Originally this was just for eng, but other teams found it so helpful we expanded it company-wide. Here are the four guiding principles: 1. You must stand behind every idea and sentence It is your responsibility to make sure that the entire document is representative of your own thoughts before you share it. 2. Writing is thinking Spending time on the writing process teaches you more about your topic. If you circumvent this process, you will walk away with a poorer understanding of the subject matter. 3. More time should be spent writing a document than consuming it If you generate a document from a short prompt then ask your readers to go through the longer output, you are disrespecting their time. They can talk to ChatGPT themselves if they want to. 4. Longer is not better AI makes it easy to generate long docs, and it loves padding them with sentences that say nothing. If you're producing docs from a short prompt, consider just sharing the prompt. You can read (and borrow) the full policy below. Hopefully we all communicate a bit more clearly now :)
179
377
221
4,185
745,090