Software is a training ground for agents
Repositories are not miniature worlds. They are useful because work produces concrete, inspectable consequences that an agent and its reviewers can learn from.
Research
Research and working papers on AI-native software infrastructure.
Field notes on runtimes, workspaces, safety boundaries, and evidence systems for AI-native software.
The runtime is the product
A capable model is only one component of useful agentic software. Task state, authority, isolation, review, and recovery determine whether work can survive contact with a real team.
From prototypes to durable systems
AI-assisted prototyping is valuable. The real work begins when an experiment leaves its maker and becomes a system other people need to understand, change, and operate.
The agent-native workbench is not a chat window
Conversation is useful for intent. A workbench for agentic software must also show what is true now: the task, its changing state, its approvals, and the evidence behind progress.
Evidence is an interface
A completion claim is useful only when a reviewer can see the checks, artifacts, and remaining uncertainty that make it worth trusting.
Capability should not decide authority
A stronger model should make better use of a defined permission set. It should not quietly acquire broader access to the systems around it.
State belongs in the system
The context window is a place to reason, not the place where a team's work should live. Durable systems keep task state, decisions, and artifacts outside the model.
The reviewable path to production
Autonomous work becomes shared progress only when a team can inspect what changed, what was verified, and what authority was used before it reaches a durable system.
Tools are security boundaries
Giving an agent access to an external system is not a prompt feature. It is an identity, credential, policy, and audit decision.
Model portability is operational resilience
Choosing a model is not a one-time product decision. Durable infrastructure keeps work, policy, and evaluation stable while models, providers, and costs change.
Long-running work needs recovery
An agent that can continue after interruption needs more than a retry loop. It needs durable task state, idempotent actions, refreshed authority, and an honest account of what is still true.
Measuring agent work in live systems
A task that looks successful in a benchmark can still create expensive review, fragile changes, or hidden operational debt. Evaluation needs to follow work into the system.
Interfaces for interrupted work
Most agent tasks will be paused, redirected, or handed off. The interface should make that ordinary reality visible instead of treating every interruption as a broken conversation.
Software that keeps a record
AI-native systems need more than generated outputs. They need a durable account of decisions, changes, evidence, and responsibility that survives the task that produced them.
What open source preserves
Open source is not a substitute for product quality or governance. It preserves the ability to inspect, adapt, and operate the infrastructure that increasingly holds a team's AI-native work.
The economics of attention in agent systems
The scarce resource in agentic work is not only compute. It is the human attention needed to make consequential decisions, inspect evidence, and recover from uncertain work.