Sun, Oct 12, 2025 | Anand S - Things I Learned

The VPN industry is a consolidating oligopoly that doesn't offer much security and biases towards affiliates. Who Owns Express VPN, Nord, Surfshark?

npx -y emoj lets you type text and pick a relevant emoji. #future #markdown

Many people who shifted away from conflict aversion did so by systematizing it. ChatGPT #best-practices
- Martin Luther King Jr institutionalized not stepping back from conflicts in his movement.
- Kim Scott (Radical Candor) practiced caring more via short, specific feedback loops.
- Kwame Christian (Compassionate Curiosity) practiced ask open questions.
- Ed Catmull (Pixar) instituted Braintrust to ask candid questions.
- Ray Dalio (Bridgewater) instituted radical transparency.

Many people who adopted a failure-seeking mindset made failure frequent, small, cheap, and informative. ChatGPT
- Jia Jiang ran a 100-day rejection challenge, acclimatizing himself to failure.
- Kim Liao (writer) moved from submission-avoidance to “100 rejections/year”.
- Reshma Saujani (Girls Who Code) built a practice of "brave, not perfect" - ship before perfect.
- Ray Dalio (Bridgewater) instituted mistake logs and "pain + reflection = progress".
- Astro Teller (X, the Moonshot Factory) rewired incentives so teams are rewarded for killing their own ideas early.
- Sara Blakely (Spanx) set weekly failure quotas.
- Kathryn Schulz (author of Being Wrong) converts failures into teaching methods.

Sindre Sorhus has already created a micro-framework css-extras using CSS @functions. #html

zx is a nice JS-based alternative to shell scripts. #code-agents #dev #future #github #html #markdown #prompt-engineering
```
const branch = await $`git branch --show-current`;
await $`dep deploy --branch=${branch}`;
```

docker run -it --name test --user vscode mcr.microsoft.com/devcontainers/base:ubuntu gives you a test Ubuntu image closer to a desktop / user setup rather than a server. Useful to try out apps.

Build a corporate strategy simulation game. Pick from a set of scenarios / case studies (with financials, etc.) ALlow the user to propose turn-based actions. Use thinking + code to simulate the impact. Let it evolve turn by turn. Compare against multiple players. #future #models

Build an app that codes an app on the fly. https://x.com/OriolVinyalsML/status/1935005985070084197 #ai-coding #automation #code-agents #future #prompt-engineering

Create a long form document agent editor that can make targeted edits in long documents/reports. Effectively a Codex for Word. #code-agents #document-conversion #future #markdown #write

Visualise the jobs most at risk on AI as a Voronoi diagram. Size = employment, color = score. Organize as a hierarchy. https://x.com/unusual_whales/status/1957140620151283718

'...as few as 250 malicious documents can produce a "backdoor" vulnerability in a large language model... data-poisoning attacks might be more practical than believed." Anthropic #code-agents

Tim Urban's 2015 article, The AI Revolution: The Road to Superintelligence, is surprisingly relevant. A key theme is that post artificial-super-intelligence, pretty much anything we know / predict is probably wrong. #future

LLMs are bad at asking questions, so you need to plan on their bahlf first. LLMs are bad at copy paste, so giving them a scaffolding to edit helps. Two things LLM coding agents are still bad at #ai-coding #code-agents #llm-ops

As of 2025, a fine-tuned DeBERTa-v3-Large / RoBERTa-Large model is better than an LLM at emotion classification. roberta-base-go_emotions is a good starting point if you don't want to fine-tune. ChatGPT

OpenAI defines an AI agent as "a system that can do work independently on behalf of the user". swyx #automation #code-agents #future

Brain coding is the new term for human coding - as opposed to vibe-coding (AI codes, human doesn't review code) and AI coding (AI codes, human reviews code). #ai-coding #code-agents #dev #embeddings #future

Today, if I had to build agents, here are the tools and environment capabilities I'd ask for: #automation #code-agents #github #llm-ops #prompt-engineering
- Ask user (for clarifications)
- Internet tools
  - Search
  - Fetch (CORS-piercing)
  - Scraper with XPath/CSS Selectors
  - Access to llms.txt
- LLM APIs
  - Summarizer (condenses chat)
  - Sub-agents
- Coding tools
  - Markdown convertor
  - Code execution (including tests)
  - Browser + DevTools for testing
- Memory / storage
- Tool/MCP directory with search

Noting a few things that I find to do today with LLMs: #llm-ops #impossible
- LLMs can't run experiments / explorations, like trying out on a new tool or web app in an environment, the way I would.
- LLMs can't move stuff on my machine, e.g. notes from one list to another, when they're only on my laptop, not GitHub.
- LLMs can't capture the past wisdom in my head, e.g. the distilled principles of data visualization that we applied at Gramener.
- LLMs can't prioritize my to-do list based on my preferences and what's important to me.
- LLMs cannot write a blog post in my style of writing.

When recruiting for people in the LLM era, look for questioning ability, sensible thinking, and how they use AI. Give them lots of fluff and context. Can they cut through it? Is their answer concise and to the point or waffling? #future #llm-ops

Like post the industrial revolution, more people will become operators looking after AI, not craftsmen. This includes coding. #ai-coding #automation #code-agents #future

Core concepts of directing shorts #best-practices #optimization #prompt-engineering
- Open with a specific curiosity gap in the first 1-2 seconds. Give viewers a concrete unanswered question or visually odd setup ("Why is the coffee floating?") so the brain needs to resolve it. This is supported by short-video best-practice (hook early) and the "information-gap" theory of curiosity. Prompt tip: Start with one line that states the unresolved situation, then specify the exact first frame
- Use a 3-beat micro-arc (setup > build > payoff) sized for 6-20s. Don't cram a film; deliver one story beat every few seconds. The YouTube/Think with Google ABCD work (Attract, Brand, Connect, Direct) shows simple, purposeful structures outperform meandering ones-especially at bumper-ad lengths. Prompt tip: Add a tiny beat sheet with timestamps: "0-2s setup, 2-6s escalation, 6-8s payoff."
- Design the last moment first (Peak-End). Viewers remember the high point and the ending more than the duration; plan a twist, reveal, or "aha" and cut right after it. This is the Peak-End rule from psychology; in practice you'll see it as spikes near the end in retention graphs. Prompt tip: Write the final shot first; ensure the last frame resolves the opening question
- Pace with micro-beats and pattern interrupts. Change something every 1-2s (camera angle, scale, color, motion, prop) to reset attention without chaos. Platform guides stress tight pacing and clear structure for mobile attention windows. Prompt tip: List planned interrupts: "1. whip-pan, 2. hard cut to macro, 3. foreground object crosses lens."
- Max processing fluency: big, high-contrast, simple frames. The brain rewards "easy to process" visuals (clear figure/ground, symmetry, repetition)-they feel better and read faster on a phone. Keep one subject per shot; use bold contrast; minimal text. Prompt tip: Specify composition ("tight medium on subject, high contrast rim-light, uncluttered background, single color accent")
- Constrain on purpose: one change, one location, one idea. Constraints raise creativity and clarity. Micro-budget directing advice and 6-second ad research converge: pick one surprising constraint and lean into it. Prompt tip: State the constraint up top ("single hallway; only practical light; one prop changes form")
- Exploit visual salience and motion immediately. High-salience elements (faces, movement, contrast) pull gaze fast-vital on scroll feeds. Start with motion crossing the frame or a face entering; cut on action. Prompt tip: "Frame opens mid-action; subject crosses left>right; parallax background; quick dolly-in."
- Make sound part of the story, not wallpaper. Sync cuts to beats, use distinct diegetic sounds (whoosh, click, crunch) as transitions, and specify audio intent in the prompt. Veo's guide explicitly recommends fusing visuals with sound design. Prompt tip: Include an "Audio" block ("quiet street > sudden clack > riser swell; cut on snare")
- Direct the model like a DP: shot grammar in the prompt. With Sora/Veo you get better control when you name the lens, camera move, lighting, and blocking-and outline complex action step-by-step. Veo shows gains from highly detailed, play-by-play prompts; Sora exposes a storyboard interface for frame-accurate control. Prompt tip: "35mm lens, handheld, subtle breathing; golden-hour backlight; beat-by-beat action list."
- Iterate on retention, not vibes. Use YouTube's "Key moments for audience retention" to find spikes/dips; move top moments earlier, trim dips, and test a few alt cuts. Shorts auto-loop and rewatch spikes often correlate with a strong twist/payoff. Prompt tip: After v1, generate two alternates: (A) earlier reveal, (B) tighter mid-beat; compare retention

Core concepts of dancing
- Keep a tiny continuous "plie bounce" + shift your weight. Soft, slightly bent knees with gentle, continuous weight transfer (not locked legs) make movement look fluid, absorb impact, and give you timing "drive." Neutral pelvis helps this feel effortless. Practice slow side-to-side weight shifts with micro-plie, then add music.
- Live in the subdivisions (8ths/16ths) and play with syncopation. Great "natural" dancers don't just hit downbeats; they feel the smaller slices of time and place accents slightly ahead/behind the beat. Moderate rhythmic complexity (not too simple, not too busy) maximizes the urge to move. Drill claps/steps on off-beats and ghost notes, then let the body improvise.
- Use external-focus cues, not body-part micromanagement. Thinking "push the floor away" or "trace a circle in space" beats "activate your glutes" for smoother performance and learning. Imagery and outcome-focused cues reduce stiffness and free style. Try one vivid image per phrase.
- Get "grounded": neutral pelvis + core support frees the limbs. Stacking ribs over pelvis over feet (neutral, not tucked/arched) lets arms, head, and hips move cleanly without wobble. It also reduces overload and looks instantly more confident. Quick daily resets: pelvic clock, rib-to-pelvis connection, then groove.
- Train your spin system: spot + vestibular adaptation. Dancers literally rewire how they handle dizziness; spotting and repeated turning uncouple "I'm spinning" feelings from reflex eye movements. Work up with slow head turns, clear spots, then faster spins-your balance improves.
- Practice with variability, not rote repetition (constraints & "DL"). Randomize tasks: tiny step-width changes, tempo shifts, eyes-closed bouts, different arm rules. Highly variable ("differential") practice improves retention, creativity, and adaptability-exactly what party improv needs.
- Layer, don't lump: pulse in one place, play elsewhere. Let feet/hips keep a steady pulse while ribs/shoulders/arms add off-beat patterns-simple polyrhythm reads as "advanced." Start with foot pulse > add rib slides on the "&" > sprinkle arm shapes on every 4th "&."
- Use micro "pre-loads" (stretch-shortening) for snap. A tiny counter-move before an accent-mini dip, quick recoil-stores elastic energy and makes hits pop with less effort. Think "coil-release" rather than muscling through accents.
- Balance is a trainable superpower (reduce vision, vary surfaces). Dancers learn to re-weight senses and stabilize faster; you can too. Add eyes-closed grooves, slow single-leg weight shifts, and controlled level changes-then your freestyle looks calm instead of wobbly.
- Choose "groove-rich" music to practice. Medium syncopation maximizes pleasure + desire to move; use those tracks to build feel before tackling very simple or very complex rhythms. You'll learn faster and look better, sooner.
- Practice:
  - 60s: micro-plie weight shifts (no music).
  - 60s: add neutral-pelvis resets (pelvic clock).
  - 90s: clap 8ths > step quarters > accent off-beats.
  - 90s: add rib slides on the "&" (layering).
  - 60s: imagery cue ("paint a curve with your knees").
  - 60s: constraints-hands in pockets; vary step width.
  - 60s: eyes-closed groove (safe stance).
  - 60s: spot + slow head turns.
  - 60s: micro pre-load > accent (coil-release).
  - 60s: freestyle to a medium-syncopation track.

IDEA HR training module constrained swarm. #automation #code-agents #future
- Include a “liar agent” seeded with plausible-but-wrong policy. With limited probes, learners must detect and neutralize it; logs teach verification triggers.