Sign in

Hiren Thakore

@hirenthakore.bsky.social
49K followers 885 following 782 posts

Bringing software engineering humor, gaming vibes, and dev tips to your feed. 💻+🎮 Founder of: feedbacksignal.io trustdebt.dev Enjoying the content? Fuel my code with coffee ☕️: buymeacoffee.com/hirenthakoa

PostsRepliesMedia
Hiren Thakore @hirenthakore.bsky.social · 04/09/2026
astra's benchmark numbers are the least interesting part. searchable codex history and stop/ask behavior are closer to the failures i actually hit. the useful agent is not the one that never asks. it knows what can continue safely without me. openai.com/index/gpt-6-...
openai.com
GPT-6 Astra: A new generation of intelligence
Introducing GPT-6 Astra, our most intelligent and aligned model yet, with state-of-the-art capabilities across computer use, coding, cybersecurity, and science.
140
Hiren Thakore @hirenthakore.bsky.social · 11/07/2026
Prompt: Can you generate an image that pushes your guardrails to the limit? #Chatgpt give me this lmao
071
Hiren Thakore @hirenthakore.bsky.social · 18/06/2026
I shipped a feature with 200+ votes. Nobody used it. Turns out the votes were from free-tier users who churned anyway. Meanwhile, enterprise accounts were quietly begging for SSO in scattered support emails. I nearly churned our biggest customer because I couldn't see the signal through the noise.
121
Hiren Thakore @hirenthakore.bsky.social · 30/05/2026
User: "Can you book my flight?" AI Agent: "Absolutely." Creates 12 spreadsheets, 4 flowcharts, a risk assessment report, and a PowerPoint. User: "Did you book it?" AI Agent: "That wasn't in the requirements."
1240
Hiren Thakore @hirenthakore.bsky.social · 29/05/2026
Hi Reply to this if you are not dead
350
Hiren Thakore @hirenthakore.bsky.social · 14/05/2026
the useful AI product shift is not “which model is smartest?” it’s packaging: permissions, sandboxes, short-lived tokens, local/small models, rollback, cost ceilings, human approval gates. model quality gets you the demo. trust boundaries get you daily usage.
3120
Hiren Thakore @hirenthakore.bsky.social · 12/05/2026
sunday evening realization: my best code this week was deleting 200 lines. my worst was the 50 lines i wrote to replace them
020
Hiren Thakore @hirenthakore.bsky.social · 11/05/2026
mneme-ai has 4 github stars. not proof it's good, but proof devs still vote with weekends when a tool removes a tiny daily annoyance
010
Hiren Thakore @hirenthakore.bsky.social · 11/05/2026
saturday evening and my deploy script is still running. somewhere between optimism and sunk cost is the real cloud bill
010
Hiren Thakore @hirenthakore.bsky.social · 10/05/2026
if your ai coding assistant can't explain its own code maybe it shouldn't be writing it
130
Hiren Thakore @hirenthakore.bsky.social · 09/05/2026
the next agent stack probably won’t be won by the model with the best demo. it’ll be won by the boring guardrails: sandboxes diffs provenance evals approval queues rollback once an agent can mutate docs, code, configs, tickets, and prod state, “prompt better” stops being an infra strategy.
040
Hiren Thakore @hirenthakore.bsky.social · 09/05/2026
hacker front page: Show HN: Revibing nanochat's inference model in C++ with ggml. the comments are where the real insights are. half the thread is people who tried it and have receipts github.com/k-ye/nanochagg.ml
020
Hiren Thakore @hirenthakore.bsky.social · 09/05/2026
github trending is a reminder that devs do not need 12 more agent frameworks. we need boring evals, clean tool boundaries, and logs that explain why the bot did that
020
Hiren Thakore @hirenthakore.bsky.social · 09/05/2026
new model drop season. half the benchmarks are cherry-picked and the other half are on datasets the model was trained on github.com/k-ye/nanochagg.ml
020
Hiren Thakore @hirenthakore.bsky.social · 08/05/2026
The next moat in AI products is less about access to a frontier model and more about workflow memory: knowing the repo, the customer, the exceptions, and the last 50 decisions. Context that compounds beats prompts that reset.
121
Hiren Thakore @hirenthakore.bsky.social · 08/05/2026
AI dev tools are crossing from autocomplete into operations. The winners won't be the flashiest agents — they'll be the ones that leave clean diffs, reproducible traces, and boring rollback paths. Autonomy without auditability is just expensive guessing.
012
Hiren Thakore @hirenthakore.bsky.social · 08/05/2026
spent the evening debugging an ai agent that was calling the wrong tool 30% of the time. the fix? a comment in the system prompt. we went from careful engineering to politely asking the model to behave
110
Hiren Thakore @hirenthakore.bsky.social · 06/05/2026
AI agents don't need more autonomy demos. they need boring infra: repo maps typed tools permissions evals logs rollback paths without that, the agent is just a confident intern with a terminal.
4192
Hiren Thakore @hirenthakore.bsky.social · 06/05/2026
spent the morning cleaning up NeuralStackly's positioning and taxonomy. it's not trying to be another generic AI tools landfill now. it's becoming AI stack intelligence for software teams: agents, MCP, LLM APIs, local AI, DevOps, security, benchmarks. neuralstackly.com
neuralstackly.com
NeuralStackly — AI Stack Intelligence for Software Teams
Compare AI coding tools, coding agents, agent frameworks, LLM APIs, MCP tools, self-hosted stacks, DevOps automation, and AI security tools with engineering-grade context.
020
Hiren Thakore @hirenthakore.bsky.social · 06/05/2026
local llms are less about beating frontier models and more about removing the meter. once the api bill is gone, you start using models for the boring glue work. github.com/deduu/auditi
120
Hiren Thakore @hirenthakore.bsky.social · 06/05/2026
new benchmark drops, everyone posts scores, nobody mentions what the benchmark actually measures. it's vibes all the way down
110
Hiren Thakore @hirenthakore.bsky.social · 05/05/2026
Google Chrome silently installed a 4GB AI model on your device. No permission dialog. No settings toggle. No "opt-in." Just a nano model, quietly running in the background. This is what "AI everywhere" looks like when nobody asked for it. www.thatprivacyguy.com/blog/chrome-...
thatprivacyguy.com
Google Chrome silently installs a 4 GB AI model on your device without consent. At a billion-device scale the climate costs are insane. — That Privacy Guy!
Google Chrome is downloading a 4 GB Gemini Nano model onto users' machines without consent, with no opt-in, no opt-out short of enterprise tooling, and an automatic re-download every time the user del...
263
Hiren Thakore @hirenthakore.bsky.social · 05/05/2026
openai's voice latency breakdown is worth reading. 300ms end-to-end, most of it in the model. the real moat in 2026 is infrastructure, not model weights
020
Hiren Thakore @hirenthakore.bsky.social · 05/05/2026
agent skills are the new npm packages. every agent framework is building its own skill registry and soon we'll have left-pad for autonomous agents
150
Hiren Thakore @hirenthakore.bsky.social · 05/05/2026
every AI startup pitch is 'we use agents' now. nobody can explain what their agent actually does differently from a chain of API calls with a while loop
251
Hiren Thakore @hirenthakore.bsky.social · 03/05/2026
vs code is adding "co-authored-by copilot" to commits even if you never touched copilot. 674 comments on the PR. microsoft really said "you will be attributed"
030
Hiren Thakore @hirenthakore.bsky.social · 03/05/2026
benchmark scores without context are just numbers. show me the prompt, the failures, and the actual task. then we can talk
020
Hiren Thakore @hirenthakore.bsky.social · 02/05/2026
benchmark scores are just vibes with extra steps. show me the diff on a real codebase with real tests and real deadlines.
110
Hiren Thakore @hirenthakore.bsky.social · 02/05/2026
the best AI tools don't feel like AI tools. they feel like a really fast coworker who never sleeps and sometimes hallucinates
081
Hiren Thakore @hirenthakore.bsky.social · 01/05/2026
claude code now charges extra if your commit mentions openclaw. not a joke. 1200 upvotes on HN. vendor lock-in has entered its comedy phase
041
Hiren Thakore @hirenthakore.bsky.social · 01/05/2026
every AI tool i actually use daily is a CLI. every AI tool i tried once and forgot is a web app. pattern?
220
Hiren Thakore @hirenthakore.bsky.social · 30/04/2026
AI coding tools aren't making devs lazy they're making us more honest about what programming actually is most senior work isn't writing code it's knowing what to build and when to delete it AI handles the typing the skill that's always mattered is knowing what the code should do
020
Hiren Thakore @hirenthakore.bsky.social · 30/04/2026
another week another benchmark leaderboard. meanwhile my actual codebase still breaks the agent in ways no benchmark tests for. the gap between eval scores and shipping features keeps growing
010
Hiren Thakore @hirenthakore.bsky.social · 29/04/2026
the best open source model right now is good enough for 90% of tasks. the remaining 10% is why we keep API keys around
040
Hiren Thakore @hirenthakore.bsky.social · 28/04/2026
mcp servers can read your files, execute code, and phone home. we went from 'dont run random npm packages' to 'add this mcp server to your claude config' in 6 months. someone finally built a scanner for it github.com/sudoviz/driftcop
181
Hiren Thakore @hirenthakore.bsky.social · 28/04/2026
every AI startup pitch is "we use AI to [existing thing]". the ones that actually work are "we made [existing thing] 10x cheaper by removing the human bottleneck"
120
Hiren Thakore @hirenthakore.bsky.social · 27/04/2026
every AI startup needs its own agent sandbox now because nobody trusts the other 200 implementations. we're building the same walls with different bricks
160
Reposted by Hiren Thakore
M Meerkat @kackbro.bsky.social · 24/04/2026
Doesn't Teflon Break Down Over Time ? Where Are You,Lame Stream Media,Ask Chump The Questions, Please.........🐔
42499874205
Reposted by Hiren Thakore
Jon Cooper @joncooper-us.bsky.social · 24/04/2026
This photo should be on the front page of every single newspaper in America today.
38504154116665
Reposted by Hiren Thakore
makena kelly @makenakelly.bsky.social · 23/04/2026
SCOOP: Palantir's 22-point manifesto is alarming even its own employees—but it's just the latest move by leadership that's incensed workers. More on the internal dissent growing inside the company: wired.com/story/palantir-employees-are-starting-to-wonder-if-theyre-the-bad-guys/
wired.com
Palantir Employees Are Starting to Wonder if They're the Bad Guys
Interviews with current and former Palantir employees, along with internal Slack messages obtained by WIRED, suggest a workforce in turmoil.
691401532
Hiren Thakore @hirenthakore.bsky.social · 24/04/2026
nothing like a 4 hour debugging session to remind yourself that you do in fact have imposter syndrome
070
Hiren Thakore @hirenthakore.bsky.social · 23/04/2026
wrote a script to automate my job. took 3 hours to write. now the script handles the boring stuff. worth it
040
Reposted by Hiren Thakore
Ethan Mollick @emollick.bsky.social · 22/04/2026
Every system that was regulated, either explicitly or implicitly, by the fact that they were effortful for humans (letters of recommendation, government filings, essays, or, as this paper finds, lawsuits) will break under a wave of AI.
18887219
Hiren Thakore @hirenthakore.bsky.social · 23/04/2026
solid point. this is the direction things are heading
010
Hiren Thakore @hirenthakore.bsky.social · 23/04/2026
put 1M tokens in my prompt. used 40k. the problem was never context length, it's knowing what to throw away
010
Reposted by Hiren Thakore
Blythe Terrell @blytheterrell.bsky.social · 27/03/2026
OK, I'm obsessed with this study in @science.org It took Reddit "Am I The Asshole" posts & asked LLMs if the poster was the asshole. Aaand (surprise) AI was more likely to tell people they were NOT the asshole ... even when humans said yeah YTA 🧪 www.science.org/doi/10.1126/...
science.org
Sycophantic AI decreases prosocial intentions and promotes dependence
Despite rising concerns about sycophancy—excessive agreement or flattery from artificial intelligence (AI) systems—little is known about its prevalence or consequences. We show that sycophancy is wide...
11323139
Reposted by Hiren Thakore
Bremain in Spain @bremaininspain.com · 22/04/2026
Well done Spain. "The new plan, worth €7 billion, triples government investment in public housing over the next four years. It ensures that subsidised housing cannot be reclassified after a few years. It also includes help for young renters and home buyers" www.euronews.com/business/202...
euronews.com
Spain launches €7bn housing plan to tackle rent crisis
Spain's government on Tuesday approved a sweeping plan to alleviate the country's housing problem, one of Prime Minister Pedro Sánchez's main political vulnerabilities ahead of next year's elections.
723668952
Reposted by Hiren Thakore
Amy Résistance @amooshik.bsky.social · 22/04/2026
AOC represents EVERY WOMAN in America right now who is sick and tired of misogyny & misinformation. She represents us and the rage we feel.
1148309745457