Sign in

Kaushik Gopal

@kau.sh
1.5K followers 507 following 434 posts

🌐 kau.sh 👨‍💻 Principal Engineer @instacart + Caper 🛒 🎧 Fragmented podcast 🤖 Google Developer Expert

PostsRepliesMedia
Kaushik Gopal @kau.sh · 02/07/2026
So what's left for engineers? Plenty. I wrote a sketch of what that might look like, in detail, here: kau.sh/blog/ai-nati...
kau.sh
What it means to be a truly AI-native software company - Kaushik Gopal's Website
Everyone wants to build a truly AI-native company. But what does that even mean? Bolting AI onto the jobs you already have is just dabbling. In this short post, I present a sketch of what that can loo...
010
Kaushik Gopal @kau.sh · 02/07/2026
- Designers don't just land up with Figma mocks: they prototype on real Android 12 devices 😜 to test how those animations and UX actually hold up - Data scientists don't just wait for real data: they run simulations up front, generate synthetic data, and build intuition around which A/B option wins
100
Kaushik Gopal @kau.sh · 02/07/2026
- QA doesn't just file bugs: they point agents at them and file the fixes too - PMs don't just ship a 70%-done PRD and wait for eng to hit the edge cases: they point agents to walk the real code, surface those cases, and flush them out with test cases
100
Kaushik Gopal @kau.sh · 02/07/2026
Everyone says they're building an AI-native company. But what does that mean? Automating what you're already doing with AI is not enough. A truly AI-native company rethinks every role and tears down the walls between them.
200
Kaushik Gopal @kau.sh · 01/07/2026
in an effort to prove that i'm not just an AI tip dispensing bot, here's a palate cleanser of one my (& the 90s') all time favorite songs, from a podcast that does some of the best song breakdowns: songexploder.net/goo-goo-dolls
songexploder.net
Episode 295: Goo Goo Dolls
"Iris"LISTEN: APPLE · SPOTIFY · AMAZON · OTHER APPSGoo Goo Dolls formed in 1986 in Buffalo, New York, and by the time 1998 rolled around, they’d already had a pretty successful
000
Kaushik Gopal @kau.sh · 30/06/2026
bonkers! even if the bar were just 80% as good as the leading models, the cost alone makes open models an easy call, especially as they build out their loops. approaching better results? the tide, 'tis a-shifting, friends...
000
Kaushik Gopal @kau.sh · 30/06/2026
if you work at a slightly larger company and you're not already playing with open models, building a sense for when to reach for which, you're going to have a lot of catching up to do soon... 3-4 times now I've liked GLM 5.2's answer more than 5.5 xhigh or Opus 4.8.
110
Kaushik Gopal @kau.sh · 25/06/2026
it’s currently specific to my workflows but no reason it has to be honestly. let me tweak it a bit and share something your way 👍 i cobbled it up using the claude articles on html effectiveness and planotator (but i like that mine is all self contained). let me try to share something
020
Kaushik Gopal @kau.sh · 25/06/2026
Lastly, a new practice that's been really working well for me is "HTML workpads" to do all my review. Each task maintains context/logs/progress etc. in a visual HTML. More in my post: kau.sh/blog/respons...
kau.sh
Responsible Loop Engineering - Kaushik Gopal's Website
Loop engineering has two shapes: one-shot loops and autonomous loops. The autonomous kind is where future software engineers will spend real time: building the boring system around the model so it can...
210
Kaushik Gopal @kau.sh · 25/06/2026
Loops need bounds (a loop with no max turns is ~= a while(true) in prod) and agents need to be told "do not absorb uncertainty." To make this all financially feasible - tier your models. Cheap local for the always-on stuff, frontier only when it counts.
100
Kaushik Gopal @kau.sh · 25/06/2026
Every loop I've built so far ended up being a little bespoke. It's hard to find a single template for all the types of loops. I think that's actually good. It's probably going to be the work software engineers have to actually spend time "engineering".
100
Kaushik Gopal @kau.sh · 25/06/2026
I spent sometime on loop engineering. Some notes... Two types of loops: 1️⃣ one-shot loops (solved) and 2️⃣ autonomous I (personally) don't think "let the agent merge" is the right bar yet. Loops are better at producing reviewable stuff you still sign off on.
120
Kaushik Gopal @kau.sh · 19/06/2026
So... i built a new font. Introducing Moxy. Moxy = Recursive + Lilex's glyphs kau.sh/blog/moxy/
080
Kaushik Gopal @kau.sh · 11/06/2026
out of curiosity- what do you like about claude code that makes it feel much superior ?
100
Kaushik Gopal @kau.sh · 09/06/2026
Really hoping open source harnesses get more traction and push the competition forward. I know I'm super happy with OpenCode + OpenChamber right now, and it'll be hard to move away from it at this point. Full post here: kau.sh/blog/opencod...
kau.sh
OpenCode power user tips - Kaushik Gopal's Website
In this post, I’d like to talk about some power user tips for OpenCode - an open source, model agnostic harness that more people should be using. Hopefully some of the advanced use cases convince you ...
130
Kaushik Gopal @kau.sh · 09/06/2026
I don't want to paint myself into any single major AI lab corner yet, so I've been kicking the tires on open source agent harnesses. opencode has become my daily driver. I took some time to write up intermediate+ user tips that show where it really shines.
220
Kaushik Gopal @kau.sh · 03/06/2026
maybe it’s the optimist in me but i feel there’s a reckoning coming where engineers are going to actually be evaluated by how effectively they use AI and it’s an actual trait companies start to look for, in interviews. tokensculpting = !tokenmaxxing
010
Kaushik Gopal @kau.sh · 02/06/2026
June update of AI model choices Kimi 2.6 has been the blow away surprise for me; often it keeps pace with GPT-Medium. Also OpenCode + OpenChamber is another sleeper hit. This harness combo is allowing me to do a lot more. Full post @ kau.sh/blog/ai-mode...
010
Kaushik Gopal @kau.sh · 26/05/2026
i know it’s getting mixed reactions but i personally love the music from #TheMandolorianAndGrogu to see how big a swing it is from the established forms (big John Willas fan as well) you should stay through the credits (also something i wish more folks did) and hear the range.
000
Kaushik Gopal @kau.sh · 07/05/2026
i'd love to share this video from my alma mater IIT Madras where the current director is giving a comp-sci lecture and it's being translated to multiple indian languages: www.youtube.com/watch?v=KNEJ... also thanks dad for the share!
youtube.com
AI-Dubbed IIT Madras Professor's Lectures in Indian Languages: Sarvam Studio
YouTube video by Sarvam
000
Kaushik Gopal @kau.sh · 07/05/2026
this is fantastic. particularly like the translating of technical words and interspersed lang detection India 🇮🇳 as folks might (not) know has ~120 distinct langs + 19500 dialects, so this is a particularly fascinating and important problem to solve there.
youtube.com
AI-Dubbed IIT Madras Professor's Lectures in Indian Languages: Sarvam Studio
YouTube video by Sarvam
100
Kaushik Gopal @kau.sh · 07/05/2026
/btw is one of the most under-rated claude code commands, that other harnesses should straight up steal
240
Kaushik Gopal @kau.sh · 05/05/2026
feels like the shape of a lot more software... Agents are the new compilers. Specs are the new code. kau.sh/blog/agent-c...
110
Kaushik Gopal @kau.sh · 05/05/2026
when generating code is cheap, the spec starts to feel more valuable. i keep thinking about my own tools: usbi went from shell → Go → Rust → .kts. Podsync went from Python → Rust. implementation changed rapidly but spec stayed the same
kau.sh
Agents are the new compilers. Specs are the new code. - Kaushik Gopal's Website
Linus Torvalds recently said1 AI will be to code what compilers were to assembly — freeing us from writing it by hand. Around the same time, I talked with Jesse Vincent (creator of one of the most pop...
160
Kaushik Gopal @kau.sh · 27/04/2026
introducing "Agent Kombat" take one prompt or plan and turns it into a planning debate between Claude Code and Codex. the agents produce independent plans, then they argue, a judge picks the stronger plan & writes the final plan you end up with a much better plan in the end kau.sh/blog/agent-k...
kau.sh
Agent Kombat - Kaushik Gopal's Website
Most multi-agent coding setups I see today look like task parallelism. You split the work, hand each piece to a different agent, and merge the results at the end. That is useful. I do it too. But I’ve...
010
Kaushik Gopal @kau.sh · 27/04/2026
Love how Mitchell said of the current AI hypsters: If there's all this great value being created with AI... where are the actual products? Full episode: Youtube: youtu.be/pKrsF3eQQnQ?... Website: fragmentedpodcast.com/episodes/310 Spotify: open.spotify.com/episode/0zht...
030
Reposted by Kaushik Gopal
iury souza @iurysouza.dev · 15/04/2026
This week @kau.sh and I had the pleasure to chat with Mitchell Hashimoto on the pod🎙️! Refreshing to hear someone of his caliber bring such a grounded perspective to agentic coding. We're super proud of this one!
151
Kaushik Gopal @kau.sh · 14/04/2026
I look up to @mitchellh.com as one of the best developer voices today.this ep. nearly broke me to edit-every 10s was a 💡 moment and I kept going "nope, can't cut that either." Can't thank him enough for being so gracious with his time.proud @fragmentedpodcast.com moment for me & @iurysouza.dev
061
Kaushik Gopal @kau.sh · 03/04/2026
Software engineers are basically becoming Harness engineers. Mitchell Hashimoto coined this term in his popular post detailing out his AI adoption journey. OpenAI then wrote a whole post with even more dtls on how they’re thinking about it and making systemic changes. kau.sh/blog/harness...
kau.sh
We are becoming Harness Engineers - Kaushik Gopal's Website
The role of a software engineer is shifting. Not toward writing more code but toward building the environment that makes agents reliable. Think about what you actually do with Claude Code or Codex tod...
050
Kaushik Gopal @kau.sh · 02/04/2026
So you’re an android dev and got 4 worktrees going, with gradle daemons roaring and frying your machine … wdyd? you go to the ☁️ with background agents ! listen to the newest episode to find out more.
050
Kaushik Gopal @kau.sh · 24/03/2026
this one was fun to make. @iurysouza.dev and I took a crack at explaining how image models (stable diff.) work in ~20 minutes. building an intuition around AI is starting to matter a whole lot and that starts by understanding the mechanics. hope you get as much out of it as we did making it!
031
Kaushik Gopal @kau.sh · 20/03/2026
Lots of "I built this with AI" posts lately. Here's one where I actually enjoyed the building part. Introducing PodSync (for editors): a podcast alignment tool (uses VAD, MFCC fingerprinting, x-correlation) AI helped me 👨‍💻 it in Rust but yrs of aligning tracks 👂helped the algo: kau.sh/blog/podsync
kau.sh
Podsync - I finally built my podcast track syncer - Kaushik Gopal's Website
I host and edit a podcast1. When recording remotely, we each record our own audio locally (I on my end, my co-host on his). The service we use (Adobe Podcast, Zoom, Skype-RIP) captures everyone togeth...
030
Kaushik Gopal @kau.sh · 19/03/2026
absolutely! this is about 95% similar to what i shared github.com/kaushikgopal...
github.com
dotfiles/.config/tmux at master · kaushikgopal/dotfiles
minimal dotfiles. Contribute to kaushikgopal/dotfiles development by creating an account on GitHub.
000
Kaushik Gopal @kau.sh · 19/03/2026
and it's even faster for me to switch windows/sessions with an fzf powered window switcher. i simply type the name of the thing i'm doing and i'm popped right in.
010
Kaushik Gopal @kau.sh · 19/03/2026
i know cmux is the rage but a highly customized tmux still feels 👑 for power users. it's quicker for me to get a sense of the sessions and windows just by looking at them in one shot (ofc they're completely rename-able) updates or alerts in the session? the panes light up.
230
Kaushik Gopal @kau.sh · 18/03/2026
Here's my list of reasons for using OpenCode 1. Switch between models on the fly 2. client-server architecture (a.k.a built-in remote control) 3. Subagent + mode features 4. Opinionated UX 5. “Highly” customizable via plugins kau.sh/blog/opencod...
070
Kaushik Gopal @kau.sh · 17/03/2026
This is the real work that teams need to put in, to get the best results from their AI coding agents. listen to our latest episode on Harness Engineering.
041
Kaushik Gopal @kau.sh · 12/03/2026
i like a lot of the new tmux-ish emulators like cmux but the smoothest workflow i've found is just using ghostty & one main tmux session. ☝️ one window for each task i'm working on with name "i" recognize quickly. then use fzf to search the tmux window list pane & switch ⚡
030
Kaushik Gopal @kau.sh · 10/03/2026
love the analogy that Iury gave - think of your AGENTS[.]md like the constitution of your codebase! we also go into an important paper that tries to answer the question that we've all had at some point - are these AGENTS files we're all meticulously creating, even worth it?
041
Kaushik Gopal @kau.sh · 07/03/2026
🎯 spot on. this is also my general take on why software engineers will always exist ☝️ but we're going to need far fewer of em to handle that 30% 🙁
000
Kaushik Gopal @kau.sh · 07/03/2026
i still actively read and curate what goes into my obsidian notes cause it's sacred to me. but i worry about losing those extra 🧠 connections that happen when you actually type or write out the notes. 🤔 on the flip side, i'm consuming & learning way more with this approach.
010
Kaushik Gopal @kau.sh · 07/03/2026
not sure it's a good thing (yet) but i've started treating my @obsidian.md vault as my (private) persistent memory. i go back and forth with an agent to research/learn, then dump "our" learnings into a note. to refresh my learning, i ask the agent to read the note and then resume
110
Kaushik Gopal @kau.sh · 03/03/2026
if you're dying to try /remote-control with claude code, you should really try opencode, which is almost the exact same thing (but imo a simpler execution) !
010
Reposted by Kaushik Gopal
iury souza @iurysouza.dev · 24/02/2026
@kau.sh and I bring you some agentic coding goodies on the pod's newsletter. Check it out!
031
Kaushik Gopal @kau.sh · 24/02/2026
This was fun for both of us! We share some of how the fragmented 🌭 is made + Tips from @iurysouza.dev and me, and a crazy way we used AI to solve very real audio problems 😉.
031
Kaushik Gopal @kau.sh · 21/02/2026
it's tempting to think of it as just an open-source Claude Code/Codex variant. it's def. more ! built on a server-client architecture, it makes hopping between mobile, desktop, and web super easy.
000
Kaushik Gopal @kau.sh · 21/02/2026
opencode is basically openclaw for coding. an interesting focus area in the world of agent orchestration is "access fluidity". how easy is it to reach the agents doing the work on your machine - from your phone, terminal, IDE, or browser? kau.sh/blog/opencod...
kau.sh
Agentic Fluidity - OpenCode is OpenClaw for coding - Kaushik Gopal's Website
One of the reasons OpenClaw got so popular was how fluidly you can chat with and operate your agents. Pull up your phone, send a quick message on WhatsApp, and you’re in business. As we focus more on ...
220
Kaushik Gopal @kau.sh · 17/02/2026
new episode is out! spot the easter 🥚? recent models have shown massive improvement owing to clever use of "modes" and subagent dispatch. it really clicked for me after chatting with @iurysouza.dev listen to improve your fundamentals! (not just tactics).
092
Reposted by Kaushik Gopal
iury souza @iurysouza.dev · 17/02/2026
🎙️Check out the new @fragmentedpodcast.com episode on subagents! They can be a huge unlock once you understand how it all works and you can get real benefits without having to a swarm them. @kau.sh and I built an RTS mental model for how they work (yes, we did that). Check it out!
062
Kaushik Gopal @kau.sh · 11/02/2026
🤫 (I use keynote with magic move to record the animations and play it as a video in the markdown 😂) 🩰
100