Reading, learning, teaching, lifting, running, ranting, software engineer, farm-owner
Joined October 2009
- Tweets3.4K
- Following617
- Followers245
- Likes47.7K
Bhashit Parikh retweeted
AWS Emulator is here!
Building API emulators has been huge for Amazon teams this year. We let agents run against dependencies over and over in a feedback loop, all in the local env. The most popular one internally emulates AWS APIs: it's 100% built by agents and now open source.
Software factory: aboutamazon.com/news/aws/how…
Blog: aws.amazon.com/blogs/aws-ins…
Source code: github.com/aws/aws-emulator
Evals: github.com/aws/aws-emulator-…
User guide: docs.aws.amazon.com/emulator…
Bhashit Parikh retweeted
Here's the problem.
If a Muslim or Christian complains about #AirPollution he is anti-Hindu.
If a Sanatani complains about it, she is a leftist or dynasty supporter.
If a tourist complains he is disrespecting Indian culture.
If a scientific historian says there were no fireworks in this world when Lord Ram returned from Vanvas he is funded by foreigners.
If a sane centrist says her kids have a breathing problem she is asked to protest against the bombs in the Iran war.
Everything is reduced to an abusive binary.
That leaves statisticians wondering if they should come up with something called the Idiots-to- Citizen ratio.
#Diwali
Bhashit Parikh retweeted
For most of my career, I used the James Mickens cybersecurity threat model. In this model, there are 3 types of threats, and different protections needed:
1. Ex-boyfriend with a vengeance. To protect from ex-boyfriend, just have a strong password, rotate every N weeks and don't write it down.
2. Criminals who want ransom. To protect from these, do the above and also upgrade and patch regularly.
3. Mossad and other state actors. If these are after you, there's basically nothing you can do. They will get what they are after.
This model no longer applies. We now live in a world where every ex-boyfriend has Mossad-level tools and easy means to use them. Repeatedly.
Bhashit Parikh retweeted
Strange for me to say but if you have an SDK and it doesn't start onboarding with this kind of "copy a prompt for your agent to set it up" it's unlikely your SDK will see rapid adoption
This is from Jev
The "best practices" for software changed very fast
Bhashit Parikh retweeted
Confluent's agent skills for Kafka are surprisingly high quality. What I love about them:
- Readable by humans too. Well organized, uses normal English and Kafka terminology, not too verbose.
- Has tons of best practices with explanations
- Has "common mistakes" section
- References the docs, does not repeat them
- Repo includes evals, so you can easily check your agent with these skills.
Bhashit Parikh retweeted
Perhaps a valuable way to evaluate software teams: What fraction of the work is solving essential complexity versus building abstractions that are fundamentally dealing with accidental complexity created by previous decisions or other teams?
Bhashit Parikh retweeted
"My IPO's gonna cure cancer"
"Yeah? My IPO's gonna cure *death* !!"
Bhashit Parikh retweeted
150 min of activity per day the rest of your life. Mix it up, most of it easy, outside as much as you can, but have fun. You’ll never have to worry about any of the stuff the longevity world is consumed with.
Your daily step count and resting heart rate may offer clues to how fast you age.
People who averaged fewer than 5,000 steps a day were more likely to show a pace of aging above 1.2 biological years per calendar year. That’s more than 14 months of biological aging in 12 months. People with a resting heart rate above 80 beats per minute were also more likely to show accelerated biological aging.
On the other hand, each additional 3,000 daily steps and a resting heart rate 10 beats per minutes lower were associated with an estimated 11 fewer days of biological aging per year.
When I spoke with Dr. Steve Horvath (@prof_horvath), we discussed how researchers use DunedinPACE to test whether lifestyle changes can slow the pace of aging. What stands out in this study is that two numbers you can check on your wrist tracked with signs of aging that researchers could detect in your DNA!
Bhashit Parikh retweeted
I can't wait for everyone to vibe their own TLS impl. And their own OS, programming language, runtime! Definitely no bugs gonna happen. Make no mistakes.
Dependencies are so over!
Bhashit Parikh retweeted
People tend to realize @Steve_Yegge was right 6-12 months later. Once I figured this out, I listen to him when he says stuff. Like this:
"Having a fit brain will be a key differentiator in the new era. Just do more of [writing], every day, to get in shape and stay in shape."
@GergelyOrosz is right. From my pov there are two reasons to write your own stuff, even if it's a struggle. First, it's fitness training for your brain. If you struggle to write, your brain is flabby. Having a fit brain will be a key differentiator in the new era. And second, we would all rather read your "bad" writing than polished AI writing. Trust me on this one. We can spot even one sentence of AI writing and it stinks. Your bad writing is fine. Just do more of it, every day, to get in shape and stay in shape.
Bhashit Parikh retweeted
One of the highlights of our Summer travels- visiting the site of the gymnasium where Aristotle taught Alexander, recently rediscovered by our friend Dr Angeliki Kottaridi
Where Aristotle Taught Alexander the Great
This morning I published my first ever article with @NatGeo on the new excavations at Mieza, which may have just uncovered the Gymnasium where Aristotle taught Alexander the Great.
Bhashit Parikh retweeted
En route to become one of my favorite stories to explain how messy the energy transition is proving:
Coal mine shifts to EV trucks to save on diesel.
📄: wsj.com/world/asia/these-tru…
Bhashit Parikh retweeted
IaC is dead.
Sorry DevOps / SREs / Cloud Engineers but you're in for the same roller coaster as Devs. AI can manage infra change better than you.
Unfortunately for IaC, we don't need code because with the actual system source code and the infra, AIs have everything needed to determine change sets / approvals and dev env lensing.
btw, Effect Oriented Programming probably will help you get there; but that is inevitable for your codebase.
Bhashit Parikh retweeted
hey sam, your mclaren is only used 5% of the time. this is a terrible allocation of scarce resources. have you considered taking the bus to work
sam: local inference is a terrible allocation of scarce resources
“a lot of people want local inference, and somehow they assume that the problem is solved now because we have open-source models, we're going to be able to run them on our computers, and so we're not going to need all the stuff in the data centre
but from a global allocation perspective, actually it's quite terrible to have everybody, like, have a bunch of the world's limited supply of DRAM sitting in their own computer and being used five percent of the time
you want maximum utilisation in the cloud if you're trying to deal with this very supply-constrained environment
somehow many people's intuition is the opposite of that”
