Dileep Patchigolla

so just to recap this week (so far) - musk industries is real (spacex, tesla, xai merger) - clawdbot explosion leading to a bankrun on mac minis but then anthropic released their own version - tesla dropped the bomb they’re halting production on model s and x to scale 1M optimus humanoid robots this year instead - china dropped the mother of all open source models kimi k2.5 that turn video into production-ready apps but then google dropped a gemini update ON THE SAME DAY that does the same thing gg - google said fuck it and also launched the worlds greatest world model genie and switched on gemini for 3.8B chrome browser users AND released alpha genome model that one-shots 1M dna base pairs for 3000 researchers across 160 countries AND teased new veo model - microsoft crushed earnings, launched a new ai chip but stock still tanked 10% because they *only* grew rev 39% - anthropic round 2X oversubbed raised to 20B 🏌️ - openai raising another $100B, 750B val 🏌️ - intel leaked they’re gonna help produce nvidias next gen feynman gpus - hello americas tsmc - a robot (built by figure) washed the dishes with zero human interaction - apple acquired stealth startup for $2B that can lip read - integrating their tech for new ai consumer airpods with cameras and mics - demis confirms google glass 2.0 coming this summer fckin hell

With most communications going through an AI filter, I believe we are going to see similar effect to what happened in architecture, logos etc - everything looking and sounding similar, with authenticity as the casuality.

PSA for a CTO, Head of AI, VP/Dir of Engineering, CXO: This is going to be one of the most important "back to work" weeks of your career. You must get your team aligned on agentic dev ASAP. If you're feeling behind or overwhelmed, here are some good reads to get you inspired 🧵

That time of the year when team members across US and India fight to reorganize their calenaders as the day light savings ends

I created videos of our AI product using * Gamma to generate slides * Granola to capture my voice as text * Claude to convert it to a coherent, concise script * Eleven labs to convert that script to a speech in the right accent for our customers 4 AI tools to demo our AI tool 😁

@AravSrinivas @comet Non AI workflows - oauth for example doesn't work half the times. I ended up switching to chrome as my default and comet as my backup browser.

What’s currently frustrating about Comet? @Comet will go through the comments here and triage issues for our team.

I love podcasts but I just wish they are shorter.. it's really tough completing 2-4 hr long podcasts

@AmandaPers50168 @svpino @oscarle_x For most parts, uv is way better than conda, just for the speed alone. Venv management is one more advantage. the toml files generated can be part of the project so recreating the venv is much easier than conda.

@svpino @oscarle_x I keep hearing this, nobody mentions conda which is what I use as much as possible, where am I here?

"pip install" is dead. Start using "uv add" instead. Thank me later.

@svpino @oscarle_x So true. I tried uv ONCE after the hype, and was hooked!!

@oscarle_x Trust me: try uv this weekend. Just for fun. You won’t believe how much better it is.

1h40 youtube video: Prompt Engineering in Python | Automatic and Programmatic Prompt Optimization | Complete Course How to code, your own, automatic prompt optimizer. How the most advanced prompt optimization tool, DSPy, works and how to fully leverage its capabilities. How to perform prompt engineering programmatically in Python, efficiently manipulating prompts with code to create powerful, maintainable, and composable AI programs. As a practical example, this AI tutorial demonstrates how to optimize the extraction of structured data from receipts, improving accuracy from 20% to 100%.

I love whatever is wrong with this viral AI companion named Merv

Used my health insurer's phone line and the AI couldn't understand my dob, my policy number or my "No" when it repeated an incorrect number - mostly because of noise in the background. Would be great if the AI PMs slow down and test their workflows

youtube.com/watch?v=i0P56P… Great breakdown of claude-code. What's really interesting is that there's no agent orchestration. It's all literally just prompt engineering. Shows how far prompt engineering alone can take, and how important it is.

I observed some really good junior devs frustrated with AI tools (code assistants, PRD generators etc), while strong managers loving them. Maybe it’s because AI still needs clear goals, guardrails, and feedback - the very things good managers already excel at?

"Windsurf reports Opus 4.1 delivers a one standard deviation improvement over Opus 4 on their junior developer benchmark, showing roughly the same performance leap as the jump from Sonnet 3.7 to Sonnet 4." anthropic.com/news/claude-op…

Why are they called wisdom teeth when half the time they don't even grow the right way? Petition to rename them Rogue Teeth

You can make typos, be vague, ask whatever, and the LLMs respond. The simplicity is an illusion that bites you hard. And that simplicity is what will likely prevent majority not learn how to use these tools properly.

I was a day trader in my prev life. My most imp learning was, trading looked incredibly simple because it's just button clicks, unlike, say, coding. But that simple clicks need to be backed by lots of prep, or you are screwed. I see the same now with these chatgpt and the likes

I read "The Claude Code Bible" an article written by the creator of Claude Code. Here are the main takeaways from the article, plus some hidden insights from testing it for hundreds of hours:

New @PerplexityComet use case: social media feed cleanup. Ex: 1. Open YouTube/X/LinkedIn 2. Open comet assistant 3. Tell it the content you don't like - could be politics, movies, whatever that's polluting your feed. Tell it to go through each post and mute/block the channel