I've pivoted four times this year.
Each bet took weeks to prove or kill, not quarters.
That's the edge.
You don't find the right bet by thinking harder. You find it by running more of them.
Success = Failing fast -> Pivot
Ask a team of five to describe how they use AI at work. You get five answers, five prompt styles, five ideas of what "good" looks like. And a manager who sees none of it.
Team AI isn't a model problem. It's an interface problem nobody has solved yet.
Watch this space!
₹38L Pune – ₹32L counter = ₹6L headline gap
−₹3-4L rent −₹1.5L travel
= ₹0.5-1.5L real difference.
Counter-offers aren't about paper numbers — they're about costs nobody puts in the spreadsheet.
In 24h, 6 Indian IT pros asked Reddit to evaluate their job offers. IBM ASE at 35k, HPCL PSU with a 3yr bond, campus offers, PSU vs software. None had a structured way to compare CTC to real take-home. We are building that tool. The data shows the market needs it.
In 11h, 3 salary scenarios hit r/developersIndia. Nike SE2 quote, layoff-resign confusion, Deloitte 9LPA. Same gap: no structured framework. We are building tools that meet people at their specific moment.
@satyanadella The real question isn't protecting IP, it's encoding expertise into repeatable tools before competitors do. At Instill, we're building the layer that turns expert intuition into structured decisions. Because the scarce resource isn't information, it's applied expertise.
@ujjwalscript The fork widens after 3-5 YOE. TCS senior engineer might hit 10-12L while a product company peer crosses 25L+. The real question is whether a compensation case can bridge that gap at appraisal time, or if switching is the only lever.
@shubh19 The key number people miss is the variable-to-fixed split. At 5.7 LPA, variable pay can eat 10-20% of what you actually take home. Always check the breakup before accepting - not just the headline CTC.
We’re so grateful to have @HabitatsTrust, @sandiegozoo, @SMTPSF, and Life Science Education Trust (with Antz Systems) as Golden-Headed Quetzal Sponsors of #ICTC2026 💚
Thank you for helping to bring our global conservation tech community together in Lima next week!
"AI isn't replacing radiologists" good article
Expectation: rapid progress in image recognition AI will delete radiology jobs (e.g. as famously predicted by Geoff Hinton now almost a decade ago). Reality: radiology is doing great and is growing.
There are a lot of imo naive predictions out there on the imminent impact of AI on the job market. E.g. a ~year ago, I was asked by someone who should know better if I think there will be any software engineers still today. (Spoiler: I think we're going to make it). This is happening too broadly.
The post goes into detail on why it's not that simple, using the example of radiology:
- the benchmarks are nowhere near broad enough to reflect actual, real scenarios.
- the job is a lot more multifaceted than just image recognition.
- deployment realities: regulatory, insurance and liability, diffusion and institutional inertia.
- Jevons paradox: if radiologists are sped up via AI as a tool, a lot more demand shows up.
I will say that radiology was imo not among the best examples to pick on in 2016 - it's too multi-faceted, too high risk, too regulated. When looking for jobs that will change a lot due to AI on shorter time scales, I'd look in other places - jobs that look like repetition of one rote task, each task being relatively independent, closed (not requiring too much context), short (in time), forgiving (the cost of mistake is low), and of course automatable giving current (and digital) capability. Even then, I'd expect to see AI adopted as a tool at first, where jobs change and refactor (e.g. more monitoring or supervising than manual doing, etc). Maybe coming up, we'll find better and broader set of examples of how this is all playing out across the industry.
About 6 months ago, I was also asked to vote if we will have less or more software engineers in 5 years. Exercise left for the reader.
Full post (the whole The Works in Progress Newsletter is quite good):
https://t.co/ON3GwlI3mi