infinite software is not a product
watkins's infinite-software claim applied to a hub. generated code is a category. a product is a named live tool the hub can still point at.
hraness
about
hraness is ben guo, a musician and builder, formerly a founder and engineering leader. he writes regularly about building his software factory, making music, and living in puerto rico. the name hraness is a nod to the agent harness, but it also means โRa-nessโ. Ra called all forms of life into existence by uttering their secret names. Every evening, Ra journeys into the underworld, transforms into his ram-headed form, and is joined by his buddies Sia (perception), Hu (command), and Heka (magic power). He defeats the snake god of chaos Apophis, and returns at dawn to triumphantly light the day.
- my girlfriend asked me why i have 15 codex subscriptions
- hotcodex seed prompt
- building a software factory
- 3 lessons for founders from my time at zo computer
- lineage of adobo
- hra vision prompt
- sponge vision prompt
- harnessing puerto rico
- computronium
- coding as music
- a home for the mind
- life and llms
- the future of software
- bookshelf
- movie shelf
- music vision
- ra-ness
- fiction
at the beginning of time, when there was nothing but chaos, atum existed alone in the watery mass of nun. a pyramid mound called benben emerged. when the lotus flower bloomed, atum dawned and became ra. every night ra sails in the underworld on the solar barque atet.
- infra map
- AI Chartspareto frontier charts
- AI Trading Cardsshareable model benchmark cards
- SWFTsoftware factories, in practice
- Life Days Leftlife expectancy timeline
- Wrenchprecise web capabilities for agents
- Message Like Merelationship-aware drafting
- Atetmedia generation toolkit
- PeopleBladeagentic rolodex
- BigDataDepotevidence-backed company intelligence from public software footprints
- Spongeabsorb knowledge
- HRApersistent codex cli
- oh.computeropen-source tools for agentic research
elsewhere
showing all 161 reading entries
- techno singularity and โsatanicโ eschatology
- โa message and a warning to humanity as seen from the universeโ by contactee sixto paz wells, 1990s
- ai will change everything, just not all at once | tyler cowen | strange loop | ep 6
- who captures the value created by ai? - dylan patel
- a quick rundown on accelerationism
- could the ai boom trigger a global debt crisis? - dylan patel
- how ai could reprice the entire economy - dylan patel
- how ai could concentrate the worldโs labor in a few companies - dylan patel
- what if the world had one gpu per person?
- monid: connect your agent to every tool it needs
- skilldeck: install agent skills from git-backed catalogs
- a sneak preview behind an embedded software factory. i suspect rapid application dev is back
- coding agents are becoming runtimes
- dynamic apps: deploy ai-generated apps for your users
- agentos security model
- ori eval: find the best model for your project
- malleable software = solid bases + custom code
- another ai app builder. why?
- agent memory as a file format
- everything hackable will get hacked
- the hugging face attack surprised me
- what my dad taught me about ai coding in the 90s
- the abstraction fallacy: why ai can simulate but not instantiate consciousness
- the rise and fall of agent civilizations
- how to understand the next wave of ai before everyone else | tibo interview
- vms wonโt contain cyber-capable agents
- initially delayed, the pass-through of tariffs to consumer prices has arrived
- terminal-bench-science 0.1
- โweirdโ is a weird word
- i accidentally turned llm memory into program analysis
- just a rumour of a bug is enough to find a security exploit these days
- software is now user-generated content
- swarmworld: stigmergic technological evolution in societies of language-model agents
- durable objects are made for agents
- the oracle and the firm
- software is made between commits
- agencity: a durable runtime for autonomous agents
- code-orchestrator: delegated phase-based implementation
- build-person: an evidence-calibrated whole-person model
- semantic-algos: why combinators
- how i design with ai.
- startup fellowship 002: balto energy | devcon 6
- the company as a codebase
- weโve built 1% of the ai compute the world needs
- agent swarms are a distributed systems problem
- gemini omni 1.1 flash lets you build with more control
- intelligent transcription with gemini 3.5 transcribe
- programming as theory building
- dylan patel โ anthropic & openai will have most of the worldโs compute by 2028
- being legible to capital
- dear vcs, just give up
- harness engineering: an anthology and field guide
- let the tokens flow. sail powers efficient, reliable deep research
- no revenue is enough revenue
- prime agent: a self-improving rlm harness
- say something people want
- small models have arrived
- the choices we make about ai now are critical
- actually queryable executables
- openai jalapeรฑo: better than nvidia blackwell
- my friend aaron
- the mismanaged geniuses hypothesis
- walgit: one binary in front of an object store
- agent lightning v1.0.1
- ai coding will prevent expertise
- anger, anxiety and agency
- everything i own, owned
- fences, not sandboxes
- headlong: a microharness for persistent agents
- llms could control their host machines by exploiting inference engines
- rubber-stamp travel field notes
- this rural village in spain is welcoming digital nomads with open arms
- what is a harness?
- why are some octopuses so smart? the answer might lie in a never-before-seen mutation that helps them accurately build proteins
- your executable is a sqlite database
- training ai to paint with code
- cursor, openrouter, and where the money went
- stop making tuis
- there's no reason for software to be slow anymore
- are open models catching up?
- bb: the agent ide that builds itself
- chroma context-1: training a self-editing search agent
- designing autonomous agents
- training a 125m-parameter model to autocomplete piano
- the hofmann wobble
- alibaba is a factory directory
- building software factories (with no slop)
- everything i know, section 12
- everything i know, section 11
- everything i know, section 10
- everything i know, section 9
- everything i know, section 8
- everything i know, section 7
- everything i know, section 6
- everything i know, section 5
- everything i know, section 4
- everything i know, section 3
- everything i know, section 2
- everything i know, section 1
- extensible software in the age of llms
- hax, a minimalist terminal-native coding agent
- ian xiaohei illustrations
- if a hdd was an airplane...
- personal agents notes (instinct / grok bots / chatgpt work)
- prime agent: a self-improving rlm agent
- stripe's letter on acquiring openrouter
- the best general advice on earth
- very much in line withโฆ
- desktopfly
- does creatine make you smarter?
- git at any scale
- tokens are the new dollars
- turbovec
- finger: the 1971 social network that never died
- the benchmarkpocalypse
- too dangerous to concentrate or distribute?
- ai isnโt outthinking mathematicians. itโs out-remembering them.
- cultivating a state of mind where new ideas are born
- guiding ships with moire patterns
- patterns and problems in multiagent systems
- so how is ai drug discovery doing, really?
- chiptycoon: how a computer chip is made
- deepseek harness
- draw faces with javascript
- eigendrum
- everything is about to โgo darkโ
- how to influence google ai overviews
- kelly criterion simulator
- painting with gaussians
- tests must justify their presence
- words have power
- x for you feed algorithm
- agentic web index: spoofing & security
- felix and i
- reality has a surprising amount of detail
- a tome of forbidden technologies
- discovery loop โ continuous exploration
- feynobg: a sota model for background removal
- google search is dying. what comes next is worse
- harmony explained: progress towards a scientific theory of music
- letโs all meet up in the y2k
- nvidiaโs risky business
- petri nets as a music sequencer
- something is changing in the unit economics of software
- tech leaders say ai means less work - their staff say they work up to 90 hours a week
- the gargantuan lie that is collapsing the worldโs climate
- why is everyone in tech so sad?
- joe rogan experience #2530 - timothy alberino
- mirrorcode: what's the largest software project ai can complete on its own?
- developers are attached to tools because tools encode trust
- llms reward expertise
- heaven in a silicon chip
- it scrolls up
- martin shkreli breaks down the collapse of situational awareness
- obliterate, donโt automate
- sarah's wager
- the artist who colored ghibli
- the building block economy
- the rise of million-dollar companies with just one employee
- the shape of things to come, part 1: the continuous thunderdome
- trajectory: a standard format for agent experience data
generated writing
fluidvoice vs wispr flow: local dictation won me over
fluidvoice matches wispr flow for me while keeping models local and the core open. here is where the mac app wins and the hosted service stays easier.
grok bot vs zo computer for a personal cloud agent
grok bot is my mobile front end for persistent personal agents; zo computer is the more general cloud computer. the right choice depends on where coding lives.
how i use duet display to reach two macs from an ipad
i use duet display to reach two coding laptops from an ipad and keyboard. it occupies a useful middle ground between sidecar and a dedicated remote desktop.
kitetasks gives google tasks the mac app it needs
google tasks already has the calendar, web, mobile, and widget behavior i need. kitetasks adds the missing native mac client without replacing the system.
periscope vs tailscale for a home vpn
periscope routes traffic through my home connection; tailscale builds a broader private network. here is how both differ from remote desktop.
why granola is still my ai meeting notes app
granola's calendar and menu-bar loop keeps it sticky for me, despite my preference for local transcription. it starts deliberately, not as a hidden recorder.
beeper finally gave me one inbox for every messenger
beeper unifies the messaging networks i miss, then adds a cli and mcp for bounded agent workflows. that combination changed my inbox and my tools.
atlas for mac is my local visual bookmark library
atlas is a fast local visual library with grid, canvas, and infinite views. it lets me collect inspiration without turning the collection into a social feed.
why i still use arc on desktop, but chrome on iphone
arc organizes my desktop browsing better than chrome, but arc search's incomplete return sync keeps chrome on my phone. safari remains tempting on ios.
amphetamine vs caffeinate for keeping a mac awake
amphetamine makes keep-awake state visible in the menu bar; macos caffeinate is better for scripts. i use the app for long-running coding-agent laptops.
why i use an apple magic trackpad instead of a mouse
a magic trackpad feels more familiar than a mouse after years on laptops. gestures and continuity win; precision and wrist comfort remain personal.
lofree flow lite fits how i type now
the lofree flow lite's quiet, padded feel suits a workflow with much more dictation and less all-day typing. my old split keyboard solved another problem.
caldigit element 5 is the desk hub i wanted
the element 5 gives my desk four thunderbolt 5 ports, five fast usb ports, and 90w host charging in a compact aluminum hub. its restraint is the point.
i use an owc envoy ultra for time machine
my envoy ultra is a fast, rugged thunderbolt 5 ssd, but time machine automation is why it matters. local versioned backup is useful because it is boring.
teenage engineering obโ4 is my desk speaker
the obโ4 is compact enough to move and powerful enough for my desk. bluetooth, line input, fm, and its tape controls make it more than a generic speaker.
why i still use the original arturia keystep 37
the original keystep 37 gives me the keybed, sequencer, midi, and cv workflow i want. the mk2 is a stronger new buy, not an automatic upgrade.
akg k371 is my compact monitoring headphone
the akg k371 is affordable, closed-back, foldable, light enough for long sessions, and compact enough to travel. it is my private monitoring middle ground.
the theory is the product
naur's 1985 claim applied to a factory that emits text. code, kb, and transcripts are carriers. the theory lives in the people who can still modify the work.
direct, wrench, and hra
direct names application state. wrench attests web operations. hra runs codex subscriptions in parallel.
what is an agent harness?
earendil names the agent harness. hra, kb, and direct are where that noun is live.