It is a testament to the voyeuristic nature of man, that wherever war breaks out we get this urge to 'monitor the situation'. And that is precisely what happened until the wee hours of the morning. This newsletter isn't a current events newsletter and there are times when writing about recruiting seems to not match with the vibe of the time so here goes - a small note not about recruiting at all.

Human society is complex. Very complex. It spans material conditions from utter penury and starvation to the unimaginable luxury. In the middle somewhere, actually much close to the top than the bottom are us - computer users. Some of us will rise to the top of the pile by judicious use of our computer skills. Some very few will fall to the bottom but mostly this would be considered an accident. The vast majority of us are going to stay here in this middle.

It's not bad here, but please be under no illusion that this middle ground - this stable employment, freedom from violence and freedom to express - is somehow a natural law, that it is undeniable or self evident. This bubble that we all enjoy is created and maintained, and when it is maintained by intelligent, forward-thinking people of integrity, this bubble within a bubble of our green Earth with blue skies and blue oceans is like a paradise.

Just like our systems have an architecture, the systems that sustain this bubble have an architecture. War then is downtime. A failure of system integrity or a breach in defences.

'Tech bros' are often accused of not studying humanities and in general it is a fair criticism. If you're an adult and enjoying the fruits of well run institutional mechanisms then it is incumbent on you to learn how they work, what keeps them running and how you will recognise when they fall into disrepair and corruption. One day it will be your generation in the driving seat. You should be prepared for that.

So the thought for this troubled Sunday is to learn more about the institutions that make the world tick. And none of this can be meaningful without a value system and like it as not, your value system will come from what you find beautiful. This then is the preparation we make in order to run our institutions and to make space for competing visions of beauty.

As the poet once said - Beauty is truth, truth beauty,—that is all. Ye know on earth, and all ye need to know..

This week I would like to highlight a rare kind of role that has come across my desk. It is not often that one finds a meaty role that isn't venture backed, still offers great compensation and hard problems to solve.

Bijli.in — Head of Engineering (NCR, up to 3Cr)

India is simultaneously installing 1 crore rooftop solar systems under PM Surya Ghar Yojana and millions of smart meters. Bijli.in is building what is essentially a financial exchange, but for energy — a P2P trading platform where your rooftop solar surplus gets sold directly to your neighbour at a market-discovered price. Backed by Hindustan Power Group and working with Nandan Nilekani's FIDE and the India Energy Stack Task Force.

They're looking for their first senior technical hire. A CTO-type who will architect this from scratch — smart meter integration across vendors and protocols, multi-state DISCOM billing, ML-driven per-household forecasting, automated trading agents, UPI-linked settlement. If you've built a marketplace or a fintech platform from zero and the phrase real-time order matching at household scale makes you lean forward rather than lean back, this is the one. 8-14 years, ex-founder/CTO preferred.

This is genuinely one of the more interesting technical problems I've seen — it sits at the intersection of fintech, IoT, ML, and energy policy, and it matters in a way that most SaaS dashboards simply don't.

Pratilipi — Senior ML Engineer (Bangalore)

Pratilipi is India's largest storytelling platform — millions of readers consuming stories in Hindi, Tamil, Telugu, Bengali, Marathi, and more. They need a senior ML engineer (6+ years, 2+ in recommendation systems) to build the systems that connect the right story to the right reader. GNNs, transformers, embeddings, multi-GPU training, production deployment at scale. If you've built recommendation pipelines for a content platform and care about making literature accessible in languages that the English-speaking tech world mostly ignores, talk to me.

And now on to the rest of the jobs.