Seasoned Engineering Leader. Rethinking AI-native product building & how teams develop software.
İstanbul, Türkiye
Joined July 2008
- Tweets13.6K
- Following452
- Followers2.3K
- Likes8.5K
Zekeriya Koç retweeted
we'll certainly see more assembly-level verification, but we should also ask why we see it mostly at places with virtually unlimited resources.
I don't think language semantics and program logics will die anytime soon, as they make program verification much, much cheaper.
Zekeriya Koç retweeted
We're still so small but I'm loving the team we've got at Superlogical. They're so talented and kind and love their craft. It's a different era of software development and our expectations and practices reflect that. I want to share a bit about how we work.
First, everyone is trusted to push to main. We do machine-aided code review pre-push and human plus more machine-aided code review post-push. Limited to no PRs.
Culturally, we've all agreed that if we ever hire someone who we DON'T trust to push directly to main, then they probably aren't the right fit for the team (so far, not an issue).
Some common sense here: if someone is stepping outside their lane or comfort zone, we'll open PRs. For example, I'm pretty trash at macOS/iOS UI, so I'll open PRs for UI-affecting functionality and ask for eyeballs/help. But for business logic and internals, I push to main.
Second, everyone can cut a release. Our release process is fully automated and when you start a release (for any platform) it triggers a Slack notification and a 15 minute countdown.
The 15 minute warning is a final gate that allows people to review plus collaborate on a changelog. During that 15 minutes, anyone in the company can pause the release rollout. After 15 minutes, the train leaves the station. You can't override the 15 minutes (except to pause). It gives everyone a chance to step in.
Third, we avoid synchronization through fast remediation rather than through planning. The server and clients on various platforms ship features as fast as they can. We use post-hoc human review and AI-assistance to identify feature mismatches across products ("Windows has support for X and Mac doesn't." or "Mac and Windows differ in their behaviors for Y in this way.").
This avoids expensive "sync" meetings and coordination. We trust that producers of features are mindful of what they're producing and we trust that adjacent product lines will self-synchronize quickly. In practice, we see feature parity across platforms within 24 to 48 hours.
All of this works on mutual trust. It ONLY works with mutual trust.
Everyone at all times has to trust that they've got each others back:
1. Push to main only works if you fix your mess, or if you don't, that someone else will quickly but that you'd do the same (and generally in equal amounts).
2. Anyone can deploy only works if you trust everyone to cut at the right time, with the right changelog, and that people will respect the 15 minute timer.
3. Sync through fast remediation only works if you trust other teams will pick up and integrate your work quickly, which only works if you would too. It also only works if you are fairly producing mismatches (e.g. not constantly).
We've got a great team. I know we're still small and that these practices will change in many ways as we scale. But this is already a markedly different practice than what I would've done 15 years ago.
In general, we are building from day 1 on a foundation of mutual respect, excellence, and trust. And it's been awesome.
Zekeriya Koç retweeted
Bunu yazan arkadaş, bu eleştiriden sonra engellemiş.
Gençler, size bu kötü sistemle nasıl başa çıkabileceğinizi göstermek yerine, sizi aciz ve zavallı hissettirerek pasifize eden söylemlerin peşine takılmak bu devirde çok kolay. Bunu yapmayın.
Zekeriya Koç retweeted
Olmadıkları biri gibi görünmek isteyen AI influencer’larımız için küçük bir ipucu:
Her “LLM bunu dahi başardı!!” haberini heyecanla paylaşmadan önce 2 dakika eleştirilerine de göz atın. Kendinizi bu kadar kolay ele vermek zorunda değilsiniz.
Zekeriya Koç retweeted
Okurken son aradığım şey bilgi. Okumazsam düşünme, kavram ve tasavvur ufkum şahit olduğum kadarıyla toplumla sınırlanır ve gündelik hayatımı idare etmemin bana yettiği bir hayat tarzına giderim ki bunu istemem. Gündelik hayatı "gerçek" sandıran bu kendine yetme yanılsamasıdır.
O işi uber duper super ai modelleriyle yapan arkadaşlar peki yazılımcı mi yoksa "sıfır yazılım bilgisiyle hede hödö" tayfa mı? Yok hepsi çok iyi yazılımcılar ise o yaptıklarına vibe coding demiyoruz. Ben demiyorum en azından.
1
8
487
Zekeriya Koç retweeted
Trying to reason with an AI doomer is as useful as trying to reason with a flat earther or a rabid Trump supporter.
Zekeriya Koç retweeted
Finally summarized what I've been sending to friends in DMs View article
Zekeriya Koç retweeted
AI ile ilgili hep aynı döngü yaşanıyor:
- Teknik temele dayanmadan AI’ı aşırı abartan bir güruh
- Mevcut modellerin yetersizliklerini göstererek karşı çıkanlar
- Modeller, eğitimleri ve araçlar ciddi ölçüde gelişiyor
- İlk güruh “ben demiştim” diye zafer turuna çıkıyor
Türkçesi, yaşam felsefem şudur: kazanmak için her yol mübahtır
Fatih Terim, Felipe Melo'yla Riera'nın kavgasını anlattı:
— Tesis müdüründen CD'yi istedim ve tek başıma seyrettim. Biraz, hafif sinirlenmişsin diyelim.
— Onu takıma izletsem şampiyonluğu bile kaybedebilirdik.
— Basın ‘Terim yollar’, ‘Terim disiplinlidir, gönderir’ yazdı. Hiçbirine kulak asmadım.
— Hayatım boyunca yaşam felsefem şudur: İnsanları kazanmak zordur, kaybetmek çok kolaydır.
— Sadece 3 gün kadro dışı bıraktım onları.
— Şöyle diyeyim biraz espriyle; sen biraz abartmışsın yani…
3
609
Yazılan kodların yüzde doksanını ai ajanlarının yazdığı yerde ai ajanlarının yüzde doksanını yazılımcılar kullanıyorsa ai yazılımcılığı öldürdü denilebilir mi? Yoksa başka mı bir şey olmaktadır?
4
373
Zekeriya Koç retweeted
Dolayısıyla vibe coding’e dair genel (“blanket”) yargılarda bulunmayı gayet meşru görüyorum.
Zekeriya Koç retweeted
AI’ın bütün bunları sizin yerinize yaptığını düşünüyorsanız, bunu nasıl doğruladığınızı da açıklamanız gerekiyor. Yazılım mühendisliği, kod yazma pratiklerinden ibaret değil. Sisteminizin hangi koşullarda doğru çalışacağını bilmek ve bunun sorumluluğunu üstlenmek de işin parçası.
Zekeriya Koç retweeted
Yıllarca kullandığınız yazılımın ikamesi bir günlük yazılımınız değil, diğer 100 kişisel projenizle birlikte yıllarca maintain etmiş olacağınız yazılım. O yıllara daha gelmediniz.
Zekeriya Koç retweeted
It is sad to see how clueless these guys are when it comes to how individuals and communities react to technology.
Noone is in need of such invasive, uncomfortable high-tech bullshit as the experience of Metaverse should have already pointed out.
Human beings react in the most unpredictable ways. We value privacy and intimacy more than efficiency.
These weirdos whom the zeitgeist catapulted into uncalled for importance have no idea.
They should hire a psychologist or an anthropologist to understand how humans behave on individual and communal level.
This they should do before pouring billions on stupid projects. (Unless, of course, they preach this without believing it, pushing an agenda to rip off the government or sth.)
BREAKING: Zuckerberg just described the near-future of computing, and it has nothing to do with screens.
His call: within 5 years, every person on Earth has a personal AI that knows their life, understands their goals, and does the work for them.
Not a chatbot. Not a search box.
An agent that is deeply, intimately embedded in how you live.
When the person building the social layer for 3 billion people says this is where we're going, it's not a prediction.
It's a roadmap.
Is the personal AI agent the most valuable product anyone can build right now?
Zekeriya Koç retweeted
I said "auto-regressive LLMs, in and of themselves, will not lead human-level AI"
That statement is still totally true.
First, the reasoning abilities of current AI systems are based non-auto-regressive search (which is what I have always advocated for). But AFAICT, they do it in token space, which is limited and inefficient. I have claimed that human-like reasoning must be a search in continuous representation space. It looks like the industry is moving towards that.
Second, the self-improvement methods, as currently practiced, only work for domains where the quality of outputs can be scored without human intervention, such as mathematics, code, and scenarios that can be simulated accurately. Not anything else. Humans and animals learn new skills way more efficiently than current RL methods.
Third, the multimodal capabilities of current AI assistants generally use separately-trained encoders (that are not LLMs). This is also what I've been advocating. Except that I think the best way to do this is with JEPA trained with self-supervised learning. The research community is clearly moving towards that (3000 papers on JEPA in just 4 years).
Fourth, if LLMs were a path to human-level AI, we would have domestic robots and Level-4 or Level-5 self-driving cars for consumers by now. And we don't. We certainly don't have cars that can learn to drive in 20 hours or practice like any teenager. We're still missing something pretty huge to claim human-level intelligence (let alone superhuman).
Sure, we now have computer systems that are impressive, very useful, and whose performance is superhuman in an increasing number of domains (coding being one of them).
But that's true of the entire history of progress in computer technology.
Lastly, there is a basic confusion about what intelligence actually is.
It is not the mere accumulation and regurgitation of existing declarative knowledge (which is essentially what LLMs do).
As Jean Piaget famously said, "intelligence is not what you know, it is what you do when you don't know."
It is your ability to solve new problem without any prior training, to act in previously-unknown scenarios, and to adapt very quickly to new situations with minimal training.
We're still far from that.
Zekeriya Koç retweeted
An astonishing number of people look at a chart produced by an entity with clear conflict of interest and don’t ask the question “how much should I trust this chart?”. Dude I don’t even trust my own charts, especially if they’re LLM-assisted, reward hacking is pervasive.
6 yaşından beri Aİ kullananlarla yarışamam elbet ama bu Jev modelini opus, astra falan gibi modellerle kıyaslamak biraz alıklık gibi geliyor. Modelin unique özellikleri üzerinden spesifik use case peşinde koşanlar en azından ilginç şeyler yapıyorlar.
1
4
511
LLM dünyasında son zamanların en "ilginç" şeylerinden Jev. Ne biletim sol üzerine astra "ilginç" değil bence. Daha süper falan ama ilginç değil. Jev ilginç gibi.
Ama orada da bu her "yeni" şeyin, şeylerin en inanılmazı olduğuna dair bir hype dalgasına binmeyen aklı başında içerik bulmak zor 🤷♂️
2
208