i feel like swe interviews could probably collapse into just 2 technical rounds:
• system design (how well can you articulate your ideas, do you actually know how to engineer something)
• onsite project to build a real thing with agents (can you translate your intent into high quality outcomes)
everything else we used to ask is no longer necessary
🗞️ if your team is too busy doing their 'normal job' to experiment with AI, you're preparing them to be replaced
ghuntley.com/replaced/
In this post, I guess this's the last time I'll say it. AI use is no longer optional if you wish to be employed. What follows is advice for current employees and hiring managers on how to hire in 2026 to source AI-first candidates.
Pavan Gupta retweeted
I don’t care about Navier-Stokes. I want to see AI solve the Voynich manuscript.
Pavan Gupta retweeted
The new Qwen 3.7 27B, running as a 17GB GGUF in LM Studio on my M5 Max laptop, just drew me the best pelican riding a bicycle I've seen from any model that runs on my laptop

ALT The pelican has the right shaped beak. The red bicycle has the correct shape of frame. The pelican's wing reaches the handlebars. It has legs on both side of the bicycle. There is a pleasing set of clouds, birds, sun, grass and shadow on the image, plus motion lines behind but not in front of the bird.
Pavan Gupta retweeted
Most engineers have never had sufficient access to real physics analysis — simulation was too expensive, slow, and specialized. @VinciPhysics vision isn’t just replacing Ansys; it’s bringing continuous physics reasoning to 100x more engineers and 1000x more simulations in a fraction fof the time.
Every major AI shift started with a new kind of world model.
Language. Vision. Code.
The next one is physics.
@saucentoss and I published a paper today defining what a real foundation model for physics has to be and why it enables Continuous Physics Reasoning.
If AI is going to help build the physical world, the bar has to be much higher.
Link to full paper: vinci4ddev.wpenginepowered.c…
Pavan Gupta retweeted
I upgraded my Claude token counter tool to compare different models and Opus 4.7 does appear to use 1.46x times the tokens for text and up to 3x the tokens for images - it's priced the same as Opus 4.6 on a per-token basis so this is actually a pretty big price bump

ALT Screenshot of a token comparison tool with an uploaded screenshot PNG image. Models to compare: claude-opus-4-7 (checked), claude-opus-4-6 (checked), claude-opus-4-5, claude-sonnet-4-6, claude-haiku-4-5. Note: "These models share the same tokenizer". Blue "Count Tokens" button. Results table — Model | Tokens | vs. lowest. claude-opus-4-7: 4,744 tokens, 3.01x (yellow badge). claude-opus-4-6: 1,578 tokens, 1.00x (green badge).
Pavan Gupta retweeted
Software horror: litellm PyPI supply chain attack.
Simple `pip install litellm` was enough to exfiltrate SSH keys, AWS/GCP/Azure creds, Kubernetes configs, git credentials, env vars (all your API keys), shell history, crypto wallets, SSL private keys, CI/CD secrets, database passwords.
LiteLLM itself has 97 million downloads per month which is already terrible, but much worse, the contagion spreads to any project that depends on litellm. For example, if you did `pip install dspy` (which depended on litellm>=1.64.0), you'd also be pwnd. Same for any other large project that depended on litellm.
Afaict the poisoned version was up for only less than ~1 hour. The attack had a bug which led to its discovery - Callum McMahon was using an MCP plugin inside Cursor that pulled in litellm as a transitive dependency. When litellm 1.82.8 installed, their machine ran out of RAM and crashed. So if the attacker didn't vibe code this attack it could have been undetected for many days or weeks.
Supply chain attacks like this are basically the scariest thing imaginable in modern software. Every time you install any depedency you could be pulling in a poisoned package anywhere deep inside its entire depedency tree. This is especially risky with large projects that might have lots and lots of dependencies. The credentials that do get stolen in each attack can then be used to take over more accounts and compromise more packages.
Classical software engineering would have you believe that dependencies are good (we're building pyramids from bricks), but imo this has to be re-evaluated, and it's why I've been so growingly averse to them, preferring to use LLMs to "yoink" functionality when it's simple enough and possible.
Pavan Gupta retweeted
Turns out you can run enormous Mixture-of-Experts on Mac hardware without fitting the whole model in RAM by streaming a subset of expert weights from SSD for each generated token - and people keep finding ways to run bigger models
Kimi 2.5 is 1T, but only 32B active so fits 96GB
I got a 1T-parameter model running locally on my MacBook Pro.
LLM: Kimi K2.5
1,026,408,232,448 params (~1.026T)
Hardware: M2 Max MacBook Pro (2023) w/ 96GB unified memory
Running on MLX with a flash-style SSD streaming path + local patching.
This is an experimental setup and I haven’t optimized speed yet, but it’s stable enough that I’ve started testing it in an autoresearch-style loop.
#LocalAI #MLX #MoE
Pavan Gupta retweeted
Dan says he's got Qwen 3.5 397B-A17B - a 209GB on disk MoE model - running on an M3 Mac at ~5.7 tokens per second using only 5.5 GB of active memory (!) by quantizing and then streaming weights from SSD (at ~17GB/s), since MoE models only use a small subset of their weights for each token
Replying to @maxdubler
And there are other real benefits such as fewer corrective maintenance requests and time spent cleaning and fixing things.
Matt Mahan is a builder's governor
We want prosperity, we're going to create jobs, and we demand an effective government that serves the people, not the nonprofit grifters and the special interests
And if you vote for Matt Mahan we will have a California that works again
an honor to host @MattMahanSJ at the YC office as he runs for governor - we need more founders in politics!
we also made him eat a giant sandwich… more on that later
Pavan Gupta retweeted
I like how Meta fucked llama 4 so badly that they aren’t even part of the conversation anymore
Pavan Gupta retweeted
🚨 BREAKING: Alibaba just handed the AI agent community a production-grade sandbox for free.
OpenSandbox is a full-stack platform for running untrusted agent code safely:
→ Unified APIs across multi-language SDKs
→ Docker and Kubernetes runtimes purpose-built for agents
→ Browser automation, VS Code desktop, and network isolation included
→ Designed for coding agents, GUI agents, evaluation, and beyond
Not a side project. Built by Alibaba. Open source.
1.5k stars (+1,100 this week). The secure agent infra you didn't have to build yourself.
Pavan Gupta retweeted
It got lost in the news firehose, but this is the craziest chart today, and maybe this decade
AI Takeover Complete: Data Center Construction Surpasses Office Construction For The First Time zerohedge.com/markets/ai-tak…
Pavan Gupta retweeted
It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the "progress as usual" way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since - the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow.
Just to give an example, over the weekend I was building a local video analysis dashboard for the cameras of my home so I wrote: “Here is the local IP and username/password of my DGX Spark. Log in, set up ssh keys, set up vLLM, download and bench Qwen3-VL, set up a server endpoint to inference videos, a basic web ui dashboard, test everything, set it up with systemd, record memory notes for yourself and write up a markdown report for me”. The agent went off for ~30 minutes, ran into multiple issues, researched solutions online, resolved them one by one, wrote the code, tested it, debugged it, set up the services, and came back with the report and it was just done. I didn’t touch anything. All of this could easily have been a weekend project just 3 months ago but today it’s something you kick off and forget about for 30 minutes.
As a result, programming is becoming unrecognizable. You’re not typing computer code into an editor like the way things were since computers were invented, that era is over. You're spinning up AI agents, giving them tasks *in English* and managing and reviewing their work in parallel. The biggest prize is in figuring out how you can keep ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions that productively manage multiple parallel Code instances for you. The leverage achievable via top tier "agentic engineering" feels very high right now.
It’s not perfect, it needs high-level direction, judgement, taste, oversight, iteration and hints and ideas. It works a lot better in some scenarios than others (e.g. especially for tasks that are well-specified and where you can verify/test functionality). The key is to build intuition to decompose the task just right to hand off the parts that work and help out around the edges. But imo, this is nowhere near "business as usual" time in software.
Pavan Gupta retweeted
Replying to @JetBlue
My flight to RDU for Monday 2/23 was cancelled and there is absolutely no option for rebooking-- chat has had me on hold for 2 DAYS and I've incurred > $600 in costs bec of this. I need to get back ASAP, I've already missed so much work/school. Pls help!!
Pavan Gupta retweeted
OpenClaw passed VSCode in GitHub stars, 2x PyTorch, 3x Claude Code.
This was not on my 2026 bingo card.
@steipete might've actually built the first one-person, billion-dollar company, staffed entirely by AI agents (Codex).
Pavan Gupta retweeted
Introducing M2.5, an open-source frontier model designed for real-world productivity.
- SOTA performance at coding (SWE-Bench Verified 80.2%), search (BrowseComp 76.3%), agentic tool-calling (BFCL 76.8%) & office work.
- Optimized for efficient execution, 37% faster at complex tasks.
- At $1 per hour with 100 tps, infinite scaling of long-horizon agents now economically possible
MiniMax Agent: agent.minimax.io
API: platform.minimax.io
CodingPlan: platform.minimax.io/subscrib…






































