Feedback systems should serve a community
Feedback systems should help communities spot problems and understand reliability, without making a marketplace needlessly hostile.
Lynn Cole / 标签
thoughts
A thread through the work.
Documentation is infrastructure
Documentation is infrastructure, whether it lives in code, a ticket, a notebook or a coffee-stained napkin. Write down what tomorrow will need.
Commit counts are not the point
My daily commit volume now rivals an earlier year of work. But the comparison revealed a bigger change: more ambitious experimentation and a different kind of work.
An early Codex workflow evaluation
An early Codex workflow evaluation finds useful planning and implementation alongside task-list friction, isolated workers and Git conflicts.
Why this is more than another Tuesday
AI is an opportunity to rethink workflows, not merely add another layer to broken ones. Ask why the system exists before streamlining it.
A name is a micro-value proposition
A product name is a micro-value proposition. A memorable conversation about naming came back to me while looking through 230 poetically named repositories.
When learning a new language stops being difficult
Go is the thirty-second language I have picked up. Learning gets faster, and I find myself missing the struggle that used to come with it.
The software you fall in love with
Some software becomes more than a workflow. It becomes an experience people love, share and want beside them at work.
Proactive agents need richer toolsets
Proactive agents need more than another assistant framework. Build richer tools around a clear problem and a specific reason to exist.
Agent safety starts with how authority is arranged
Agent safety starts with where authority, secrets, backups and deployment permissions live. An agent should never have enough loose authority to destroy the company.
Remembering CodeCharge’s global and local views
CodeCharge’s global, layout and data views made a memorable model for generating applications. I miss that way of seeing the work.
Try a better process before another training run
Before paying for another training run, inspect the process: clean inputs, explicit steps and properly labeled tool results solve a different class of problem.
MCP: usability, abstraction, and scaling costs
My critique of MCP’s abstraction, security and scaling costs, and why some of its use cases remind me of FastAPI.
Cleaner signals before larger models
Clean data, explicit structure and useful scaffolding matter before larger models. Better weights are not a substitute for a legible task.
Do not make your product less useful
Don't ever do anything to intentionally make your product less useful, especially when there are open source implementations of the same concept or competing products in the same space.
Radio echoes of a civilization
What if distant listeners were more interested in our phone traffic, angry emails and love letters than our broadcasts? Radio carries an intimate echo of civilization.
Recognizing my signature in Flux
Recognizing my signature in Flux got me thinking about training, adversarial art and the influence artists leave inside models.
Cheap software still needs someone behind it
Cheap software still needs support, proven behavior and someone accountable when it fails. Reconstructing a mature product remains real engineering work.
An early Antigravity workflow review
An early Antigravity review of limits, model behavior, coding competence and the difference between implementation and brainstorming.
Recursive improvement needs an external verification loop
Recursive improvement changes when an agent has tools, environmental feedback and an external verifier. The systems layer is part of the loop.
Belief is not knowledge: on technology forecasts
Tech CEOs are not prophets. Belief, suspicion and hope are not knowledge, even when a product narrative needs them to be.
A hypothesis about model behavior and prompting
A hypothesis about changing model behavior: cross-modal reasoning affects how agents code, especially when the task asks them to reason against unusual priors.
A shared study-guide system for agents
What if agents shared dense study notes the way students do? One agent does the reading, and a whole peer group benefits.
Meta-harnesses and self-customizing software
Meta-harnesses are self-customizing software. I have been building them since my first LLM agent project and want to share where the difficult pieces live.
What commit insertions and deletions might tell you
Commit insertions and deletions can reveal churn, unfinished work or oversized commits. I look for patterns rather than treating volume as achievement.
The word agent is losing its precision
The word agent now covers too many different kinds of systems. More precise terms would make the work easier to discuss.
Thinking sideways about hyper-connections
A way of thinking about manifold-constrained hyper-connections: richer sideways information flow, with geometric guardrails that preserve stability.
Why contract automation is not an easy problem
Contract automation is not just summarization. The hard work is turning complex agreements into operational rules that can be checked over time.
Digital painting needs to get its hands dirty
I want digital painting to feel less sanitary: a canvas driven by hand movements, voice commands and a game engine, where the colors can finally get their hands dirty.
Software, social capital, and the failure of clone sites
I spent two years building clone-site software. The missing ingredient was the trust, relationships, and audience around it.
You can spot art by the way it behaves
Framing, invisible sculpture, and the difference between making an image and making something that behaves like art.