1,023 karma · joined March 7, 2015
In a large enough org that experience doesn’t happen though - you have to go through and understand how the org’s infra-as-code repo works, where to make your change, and get approval for that.
Now it seems things are swinging back the other direction and articles like "Use One Big Server" are getting re-discussed: https://news.ycombinator.com/item?id=45085029
Makes having good tests even more important. One technique I've found super helpful for coding with agents is to make the agent do TDD.
Basically ask the agent to come up with the test cases first, manually review those to make sure they make sense, then have the agent game itself to write code to pass the tests. I feel like doing TDD on my own manually is very tedious but having it be AI-assisted helps me move a lot faster.
I remember browsers used to have a native RSS button in the main interface and then you could curate your feed. Seems better than any news feed thing gamified to steal my attention. Sigh.
old-man-yells-at-cloud.gif
Aaand that's startup founder life :)
Intense multitasking, needing to accept a lower engineering quality bar, and ignoring scale problems because you don't know if anyone will actually buy the thing you're building yet.
Engineering something that you know you'll redo in 1 month is very different from engineering something that you intend to last for 5+ years, but it's still a fun challenge picking the right tradeoffs and working under different constraints.
> One panelist shared a personal story that crystallized the challenge: his wife refuses to let him use Tesla’s autopilot. Why? Not because it doesn’t work, but because she doesn’t trust it.
> Trust isn’t about raw capability, it’s about consistent, explainable, auditable behavior.
> One panelist described asking ChatGPT for family movie recommendations, only to have it respond with suggestions tailored to his children by name, Claire and Brandon. His reaction? “I don’t like this answer. Why do you know my son and my girl so much? Don’t touch my privacy.”
But in practice, games that I’m into take 20+ hours to finish, and with my busy schedule, a $60 game gets stretched out to a year or so, so GamePass’ $15 fee (which is a lot higher now!) becomes too expensive
It depends on the shape of your data. In my domain (cloud security), there are many many entities and it's very valuable to map out how they relate to each other.
For example, we often want to answer a question like: “Which publicly exposed EC2 instances are used by IAM roles that have administrative privileges in my AWS account?”
To answer the question, you need to: 1. Join ec2 instances to security groups to IP rules to IP ranges to find network exposure paths to the open internet. 2. Join the instances to their instance profiles, to their roles. 3. Join the IAM roles to their role policies to determine which have admin policies. 4. Chain all of those joins together, possibly with recursive queries if there are indirect relationships (e.g., role assumption chains).
That’s a lot of joins, and the SQL query would get both heavy and hard to maintain.
In graph this query looks something like
match (i:EC2Instance)--(sg:EC2SecurityGroup)--(r:IPPermissionInbound{action:"Allow"})--(rng:IPRange{id:"0.0.0.0/0"}) match (i)--(r:AWSRole)--(p:AWSPolicy)--(stmt:AWSPolicyStatement{effect:"Allow", resource:"*"}) return i.id as instance_id, r.name as role_name
To answer this question what internet open compute instances can act as admins in our environment, we needed to traverse multiple objects, but the shape of the answer is pretty simple: just a list of ids and names.
Graph databases have quirks and add complexity of their own. If your domain isn't this edge heavy, you're probably better off with Postgres, but for our use-case it's been worth the trade-off imo.
The researcher has gotten actual shells on oai machines before via prompt injection
> Build it and they will come is a fallacy. You have to tell people about the damn thing.
Great lesson for engineering types
(side note in all seriousness I did enjoy the article)
Yes, Demon's Souls was hard, but eventually I somehow I started passing dungeons and beating bosses. The rush that I got from that gave me what I needed mentally to persevere through my classes: by the end of the semester I had A+s in 2 and an A in the other. I don't think I've had a better semester since. Beating big demons in video games made me feel like I could beat my own big demons in real life.
Lots of others feel the same way about Souls-like games; there are many video essays on Youtube that cover how Souls-likes got them through depression and other things.
There are many ways for a company to justify leaving a posting up if sued:
- candidate’s resume did not meet bar so company did not interview
- candidate could not be scheduled
- candidate was interviewed but did not pass
- offer was given to competing candidate but competitor rejected so company couldn’t fill position and still had to leave posting up
Companies would then need to keep super detailed records on job postings which is overhead, and then many would just choose to not publicly post to avoid the hassle
Telling people to “raise the bar”, and then suggesting they should aim for less engagement just sounds strange. Who decides where the bar is, and why should hitting it require fewer people to see what you made?
I agree that a lot of LinkedIn content is cookie-cutter garbage, but advising others to invest serious effort into thoughtful writing with the expectation of no reach feels deeply unhelpful.
Instead, you could encourage others to write content on LinkedIn that sounds less "what X taught me about B2B sales" and improve the culture of "toxic mediocrity".