HNHacker News
TopNewBestAskShowJobs

yaseer

1,886 karma · joined January 2, 2015

Co-founder axiom.ai, YC W21 https://axiom.ai/

Outside of work, I'm extremely interested in automated theorem proving, the lambda calculus and its applications to computational biology.

An out-of-date blog explaining my old research is here: https://yaseersheriff.github.io/

submissionscomments
yaseer··on An automatic theorem proving project
I agree he seems quite disinterested in 'machine-oriented ATP'.

But he's a mathematician, not an AI researcher - they have different goals.

Many AI researchers are interested in creating tools that solve problems (or proofs in this case).

Most mathematicians find proofs interesting for the insights they provide, not just the solution output.

You could say Gowers is looking for meta-proofs that provide insights on proof generation. It makes sense for him to emphasise the symbolic logic approach of GOFAI, here.

yaseer··on An automatic theorem proving project
Interesting, most theorem provers like lean [1], coq [2] and Isabelle [2] are concerned with formalization, and interactive theorem proving.

I initially thought "why do we need another one of these", like rolling your eyes at another programming language or JS framework. There's even another large-scale research project already underway at Cambridge [3].

I thought this extract from Gower's longer manifesto [4] captured the problem they're trying to solve:

"(just consider statements of the form “this Turing machine halts”), and therefore that it cannot be solved by any algorithm. And yet human mathematicians have considerable success with solving pretty complicated looking instances of this problem. How can this be?"

The heart of this project is understanding this problem - not creating another proof formalisation language. To understand algorithms that can prune the search space to generate proofs, using a "Good Old Fashioned AI" (GOFAI) approach, rather than machine learning.

Gowers makes a point to contrast their GOFAI approach with ML - they're interested in producing mathematical insights, not black-box software.

[0] https://leanprover.github.io/

[1] https://coq.inria.fr/

[2] https://isabelle.in.tum.de/

[3] https://www.cl.cam.ac.uk/~lp15/Grants/Alexandria/

[4] https://drive.google.com/file/d/1-FFa6nMVg18m1zPtoAQrFalwpx2...

yaseer··on Managing mental health while running a startup
Strongly agree with this. A lot of founders make the mistake of prematurely scaling, and I've done that in the past.

Scaling from a handful of people, to more than a dozen creates a significant jump in stress, and dilutes focus, for a non-commensurate jump in productivity. Any pre-PMF startup should be as small as humanly possible.

yaseer··on Marriage Markets (and Muscularity)
That's interesting, you're right, the article does report a linear-only relationship!

They may not have sampled muscular BMIs much above 30 though:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5745404/

Gym sample: > n = 61, mean BMI: 24.0, s.d.: 2.75

~5% would be above 2 s.d => 3 people with BMI above 29.5

My 'theory' would be for muscular builds with BMIs above 30, the linear relationship breaks down for heterosexual women. You might need a larger sample to test that, perhaps with a bodybuilding sample rather than college students.

yaseer··on Marriage Markets (and Muscularity)
My anecdotal experience is that the response-curve to muscularity for heterosexual women is non-linear (now there's a sentence I'd only write on HN...), and there's only a correlative effect for normal range BMIs.

I've been thin, athletic and quite muscular (BMI > 30, 12% bodyfat from powerlifting), and muscular physiques with BMI > 30 are considered less attractive than athletic physiques (Defined as BMI 20-25, 10-15% bodyfat).

Also, deviating away from heteronormative perspectives, I would speculate that the response curve for homosexual men may be linear for higher ranges.

i.e. Gay men like significant muscle, but straight women prefer athletic builds, on average. A cursory glance at some adult sites for these markets might support this admittedly anecdotal assertion.

yaseer··on Why are some egg yolks so orange?
"I can readily tell the difference between Clarence court eggs and copycats..." is an extract from 'Middle Class British Problems' if ever I heard one! :)
yaseer··on Ask HN: What do you do to recharge after a day of work?
I find it starts off this way, when your body isn't in the rhythm of exercise. As you exercise more your body changes, and its response to exercise changes.

When I was exercising 5x a week, for months, the exercise routine was energising.

When I did nothing during months of COVID, I gained weight, felt lethargic and found exercise draining.

yaseer··on Scaling lessons learned at Dropbox (2012)
This may be from 2012, but it's surprising just how much of the specific advice still applies.

Grep, sed, awk, xargs etc are still the canonical list of shell tools for debugging, and have been for decades.

Makes you realise, learning Unix tools will likely be a better long term time investment than the latest trendy framework. I wish I could go back in time and tell that to my younger self...

yaseer··on The YC Winter 2022 Batch
Would you mind sharing which companies these were?
yaseer··on Modern PHP
This sounds like an objection based on a personal principle.

I think if you tried explaining your design choices like that in a technical interview, or to a teammate, they may not be persuaded.

yaseer··on My biggest mistake as an RPA developer
Thanks!

We think UI automation has a lot of potential as a programming paradigm accessible to non-coders. Once RPA hype and noise dies down, perhaps a few signals like that will remain.

yaseer··on My biggest mistake as an RPA developer
Yep, RPA is just User Interface Automation.

Unfortunately, A lot of RPA companies, including UiPath, have co-opted the word RPA into enterprise marketing, alongside words like 'AI', so the words lose meaning. But RPA is just automating the UI, when no API exists.

Disclosure: I'm a co-founder in an RPA startup. I think RPA is a confusing, historical accident of a name.

yaseer··on Today is Y Combinator's 17th birthday
> we should acknowledge that it embraces and encourages the high-risk/high-reward model, which should be entered into with open eyes

We were W21. I would argue one of YC's main goals is to reduce the risk to founders, and enable starting startup more accessible to those with the right skills, regardless of circumstance.

Getting into YC guarantees a certain minimum amount of funding and essential support, which not everyone outside SV has access to. It's commonplace to see people quit high paying jobs, and start businesses whilst married with children (the old narrative of 20-something male dropout is no longer the norm). This all happens because YC creates a platform that reduces risk for founders.

yaseer··on Vitamin D supplements lower risk of autoimmune disease, researchers say
Yes, I think what you're describing agrees with my family's experience.

When I said 'lethargic'->'increase in energy levels', I didn't mean increase above the baseline. I meant 'lethargic'->'normal'.

yaseer··on Vitamin D supplements lower risk of autoimmune disease, researchers say
Pre supplements: Feeling lethargic and often sick, not shaking off coughs.

Our Doctor diagnosed Vitamin D deficiency.

Post supplementation (and D levels returning to normal), sick far less often with lower severity + increase in energy levels [to normal]

EDIT: Clarified energy increase is to normal.

yaseer··on Vitamin D supplements lower risk of autoimmune disease, researchers say
Vitamin D supplements are one of the few where there's a strong evidence base, and where health services advise on supplementation.

https://www.nhs.uk/conditions/vitamins-and-minerals/vitamin-...

It's particularly important if you have dark skin and live in the northern hemisphere. Several of my family were diagnosed with Vitamin D deficiency and had a tangible increase in quality of life from supplements.

yaseer··on Norway Chess cancels Alexander Grischuk
Yeah that really sucks.

It's understandable when the lines between state and individual are blurred - like with state sponsored doping of Olympic athletes, who are competing as part of a national organisation.

But when it's an individual whose nationality just happens to be Russian?

Seems more unfair, than just.

yaseer··on Increased error rate on Gov.uk on 5 January 2022
A pretty decent incident report by the standards of government websites.

As a business + personal user over 15 years, I've noticed The UK's gov.uk websites have been improving at the fastest rate I've seen over the last 3-4 years in particular.

Does anyone know why this is? (Or perhaps you have had a different experience).

yaseer··on There was a time before Mathematica (2013)
Yes, whilst I admire Wolfram's work, he combines hubris (e.g. Claiming to reinvent science[0]), with a tendency to ignore the work of others [1].

Hubris aside, he's not a conventional thinker, and I found his personal setup interesting [2].

[0] https://www.wolframscience.com/nks/

[1] https://en.wikipedia.org/wiki/A_New_Kind_of_Science#Original...

[2] https://writings.stephenwolfram.com/2019/02/seeking-the-prod...

yaseer··on Sonic Robo Blast 2: a 3D open-source Sonic the Hedgehog fangame built on Doom
I agree that's the norm, however, Sega's attitude to Sonic fangames seems to be the exception that proves the rule.

They are surprisingly receptive to the community, and in the case of Christian Whitehead, began to contract him for ports, and then eventually a Sega-approved, fan-made game; Sonic Mania.

[0] https://kotaku.com/sega-takes-shot-at-nintendo-encourages-fa...

[1] https://screenrant.com/sonic-hedgehog-fan-games-sega-respons...

[2] https://en.wikipedia.org/wiki/Christian_Whitehead

yaseer··on The Goat Boy Rises (1993)
Yep, he's got iconic status in the UK for those of a certain age.

It's also quite shocking for those of that age to look back and realise he died at 32!

yaseer··on Poll: Where do you live?
Is HN easily accessible in China?

Noticed how small the representation is in comparison to India and Japan and was curious if this is the local market's choice, or the great firewall in action.

yaseer··on Poll: Is the leetcode grind necessary to land a high paying remote job?
> Do you mean expensive in terms of taxes, the market commending a higher salary, or other reasons?

The market. The US market for high-salaried knowledge work always pays more, whether is Tech, Finance, Legal etc.

In general, if you're a high-earner/rich, America is a great place to be. If you're poor, the UK offers more support and protections, but this is a digression.

yaseer··on Poll: Is the leetcode grind necessary to land a high paying remote job?
If you live in a large city, like London, yes. But for a remote job, you're no longer tied to doing that.

There's obviously this phenomenon of high-salaried tech workers moving farther out of high cost of living places like London and San Francisco.

yaseer··on Poll: Is the leetcode grind necessary to land a high paying remote job?
It's much more expensive to hire in the US than UK, for engineering in particular.

Context - We have a UK company with US parent, and choose to hire in the UK for this reason.

yaseer··on Poll: Is the leetcode grind necessary to land a high paying remote job?
I guess it depends on the stage, and funding level.

Those that have raised a large seed round (e.g. YC companies), or Series A, $100K USD for a senior engineer is pretty common.

Most of the remote jobs advertised here will be expecting that, including the UK companies:

https://www.workatastartup.com/jobs/l/engineering

Disclosure: YC Company in London. Not hiring, but our peers are hiring with these kind of salaries in mind. Not happily mind you - employers never like it when the market favours job-seekers!

yaseer··on Poll: Is the leetcode grind necessary to land a high paying remote job?
Startups are awash with capital at the moment, and remote is the rule rather than exception, as it was 3 years ago.

Plenty of startups don't have leetcode-type coding interviews in their culture, and the job market is such that $100k remote salaries are widespread.

No, the leetcode grind is not necessary in this job market, particularly if you apply to startups rather than FAANGs.

yaseer··on Bolt founder on Stripe/YC
"i'm glad we're finally talking about this because the first time i met patrick collison he said "hey, nice to meet you, tell me about your startup" and then halfway through my pitch he waved over his guy silently and they busted my knee caps with a tire iron.

...his guys were paul graham and roelof"

https://twitter.com/nealkhosla/status/1486024266252390400

yaseer··on Bolt founder on Stripe/YC
Human beings often perceive patterns and deliberate action, where systemic explanations would suffice: https://en.wikipedia.org/wiki/Apophenia

A good example is popular discourse on 'elites'. You could perceive this as a shadowy cabal consciously acting, or as systemic flaws in society which perpetuate inequality.

Ryan's account leans towards the former, shadowy Cabal. VCs have systemic flaws in perception and they're subject to a herd mentality (e.g. startups are bimodally hot or not).

The systemic theory can explain Ryan's fundraising experiences more simply than the 'Mob Bosses of Stripe and YC' theory.

yaseer··on Ask HN: Has anyone gone from software to physical engineering?
I've done the reverse - I would recommend studying engineering, but I would not recommend the job market for physical engineering.

I studied Engineering Science with Computer Science, did some Electronic Engineering before focusing on software.

Studying diverse fields of engineering is intellectually stimulating and creates lots of transferrable skills (Who knew all that math I learned makes learning ML easy?). You will be open to jobs both in physical and software engineering + It's harder to be self-taught in physical engineering.

..Just be aware the job market for physical engineering is orders of magnitude smaller than software engineering. You will have less choice and lower pay. Most of the smart engineers I know with degrees from good schools ended up in software.

← PreviousPage 4 of 14Next →