accelerated learning

Joined March 2016
107 Photos and videos
Seeing about 5 flows of agentic coding: 1. Standard: Codex / CC chat 2. Parallel work trees: Conductor 3. Goal based: /goal 4. Ticket / Linear : OpenAI Symphony 5. Self improving Any more ?
2
1
134
Robots I saw in Shenzhen:
2
2
32
2,822
Went to Shenzhen on a whim and main mission was to see as many consumer robots out there. I’m sure companies like Unitree are cooking up more interesting things that will come to market. Started a research piece on this so will share shortly.
1
1
61
Tbh was quite inspired from these QTs. Ty for the inspo @0xTP91 @LaserDing @jesselyu @Robo_Tuo @jimchang Credit and source: x.com/zacharyvalles/status/2…

72 hours after YC demo day, I moved to Shenzhen for 8 weeks 🀠 I'm headed back to SF with new hardware in hand (sharing more soon), but some takeaways documented below: > If you have even the slightest ambition to found a hardware company, visit SZ. Pre-raise, pre-team, pre-idea, pre-job departure, it doesn't matter. Just go. > Plan your visit according to a major conference that interests you. Use that conference as a supplier meeting springboard - that's your ticket to any factory under the sun. > At the factories, ask about lead times, don't ask about cost (wait on this). Your iteration rate is driven by the lead time on the longest lead time item in your assembly. It pays to identify these parts early to build project timelines. > Visit Huaqiangbei (read: this is a mini-city, not a building). Robotic subassemblies, batteries, chassis's, electronic parts. They all have buildings where vendors are tightly clustered. Plan to spend 4-6 hours walking around before you find exactly what you're interested in. > Business relationships are valuable commodities. Treat them as such. Pay attention to people, learn about them. Bring thoughtful gifts. Wait for them to sit first. With Baiju, fill the glass but with tea leave some room. Cultural customs are fun to learn, but also convey a seriousness towards the working relationship. > Suppliers fit cleanly into discrete buckets. Level of complexity and execution on past projects indicates what is in scope for them. Trivial, but important to level your build expectations. It is easy to design a part with 12 subsequent manufacturing processes, exceptionally hard to find a supplier to fill this order. If you need coffeeshop recs, food recs, or hotel recs I have a few. Move to Shenzhen! Get to building!
2
106
Tom πŸ‘“ retweeted
Jun 1
just a small zoom out on the vibe shift: in Feb 2025 @soumithchintala was talking about his dream of personal, local, private agents, most people didn't believe him. it's June 2026 and @pewdiepie has just released his vibecoded @opencode wrapper that is a complete personal AI productivity suite including email, docs, and calendar. top of HN, easily >1m views, >10k stars in a day. if your Knowledge Work Agents startup can't beat pewdiepie you might as well pack up and go home at this point, his is the benchmark for what you can DIY.
60
12
270
32,050
codex 5.5 is gud but a hungry little token monster.
1
94
one engineer should be able to run 10 agents in parallel.
1
88
Tom πŸ‘“ retweeted
πŸš€ DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. πŸ”Ή DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. πŸ”Ή DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice. Try it now at chat.deepseek.com via Expert Mode / Instant Mode. API is updated & available today! πŸ“„ Tech Report: huggingface.co/deepseek-ai/D… πŸ€— Open Weights: huggingface.co/collections/d… 1/n
1,650
7,637
45,742
9,883,399
- Still driving with Codex app - 2 repos x 2-5 agents is my human limit - Losing joy for coding as I'm just becoming a review slave. Focusing more on distribution.
Current stack - Main: Codex app - Models: GPT 5.3 Opus 4.5 Kimi 2.5 - Background Agents in Slack: Claude Codex
1
4
374
Rooting for open source LLMs to win in the long game
2
156
openclaw is now making lofi music πŸ˜…
4
176
someone is probably going to solve .local --> .cloud agents soon. wake up -> code on computer go to sleep -> hand over context run a cloud agent
1
2
203
has anyone here been able to consistently max out their AI token usage on a MAX plan?
1
262
gg. great tool
I built markdown.new Put markdown.new/ before any URL β†’ get clean Markdown back. Cloudflare's Markdown for Agents is great, but only works for enabled sites. markdown.new works for ANY website on the internet. 80% fewer tokens. Also converts PDFs, images, audio. Free. No signup. markdown.new
2
239
marie kondo but in sf male persona.
Feb 12
We have a special thing launching to Codex users on the Pro plan later today. It sparks joy for me. I think you are going to love it...
328
Current stack - Main: Codex app - Models: GPT 5.3 Opus 4.5 Kimi 2.5 - Background Agents in Slack: Claude Codex
18 Dec 2025
Most fun I’ve had making software: - Claude Code Max Plan - Claude Skills - composer for small tasks
2
3
720
starting to think this is true the engineering moat is disappearing. now its more about fundamentals prompting seeing where CCode or Codex is going wrong. we might have to start picking up a complementary skillset (i.e. design / pm / marketing / sales) to stand out.
for anyone thinking of career building in the AI era: the T-shaped career is dead, E shaped is the new model T = broad knowledge one deep skill E = broad knowledge multiple deep skills AI can beat you in a single domain, but it can't replicate a unique intersection of 3-4 deep skills that you have instead of framing yourself as just a designer, AI is allowing you to pick up adjacent skills like "someone who also understands fintech, user research AND has shipped 3 compliance products" not just an "engineer" but "the person who can architect systems, also design the interface and talk to customers" the more specific your combination, the harder you are to replace stack skills deliberately, don't go deeper in just one thing, but pick 3-4 adjacent things and make AI be your tutor and synthesize between them still learn to code, not to be a coder necessarily but to prompt and direct code the people who thrive will be non-fungible, fluent with AI and focused on judgement the people who struggle will be generic specialists hoping their titles will be enough
2
1
2
233