Parallel source trees are cheap; parallel runtimes are not. How a disposable, fully isolated platform environment per working tree lets multiple problems — worked by humans or agents — run side by side on one laptop.
Read more →How we're building the free and open source building-blocks for the future of AI data
Read more →Dreams sound mystical. In practice, they are a way to label agent traces, reflect across sessions, and turn repeated behavior into something the next run can use.
Read more →I audited our agent sessions expecting to find wasteful repeat calls. Repeats were only 4%. The real money was in agents re-reading what they already knew, plus a permission-check tax we could finally put a price on.
Read more →What we see our AI agents doing may not be exactly what's going on under the hood.
Read more →Your team keeps paying for the same solved problem, and because the tickets all close, nobody sees it happening. Repetition is evidence of a missing team capability, and standup is the detection layer that's been announcing it all along.
Read more →London's bridges are failing under load nobody modeled. Haussmann rebuilt Paris for traffic that hadn't arrived yet. Agent infrastructure faces the same choice: skills, shared team sessions, and reflections are how you build for the traffic that is coming.
Read more →Before companies use AI cost overruns to justify cutting people, they should terminate the agent workflows that consume money without producing reliable outcomes.
Read more →One session, $100.66, 91% Fable 5 — for confirming rollout steps. Multiply that across 200 engineers and you have a budget conversation nobody is ready for. The paper console makes the numbers visible.
Read more →Evidence gets a skill written. Evaluation keeps it worth keeping. A skill that isn't measured can silently make your agent worse, and, because skills execute, a wrong one does more damage than a wrong doc ever could. Here's why every skill needs an eval, and what a real one looks like.
Read more →Skills are about to become part of the AI engineering stack. If they don't carry provenance, we're just rebuilding prompt libraries with nicer names. A prompt starts with intention. A skill should start with evidence.
Read more →Frontier labs are paying a billion dollars a year for the same kind of session reasoning your team generates and deletes every night. The asset isn't the code the agent produced. It's the path the agent took.
Read more →AI agent workflows are no longer single requests. They are stateful, branching systems that mutate their own context over time. Request logs, traces, transcripts, and bills flatten that structure. Teams need telemetry primitives that preserve session lineage, cache topology, and prompt shape.
Read more →Paper Compute's platform spans 25 repositories across 6 layers. Here's how we built a synthetic monorepo from git worktrees so coding agents stay coherent across the whole system — without Bazel, submodules, or a one-quarter migration.
Read more →We built Paper Compute to close the Agent Gap — the inability to prove what an agent did, audit its actions, or replay its decisions. Here's what we built and why.
Read more →How we built tapes to be a durable, cryptographically secure agent dataset
Read more →I hit the same Confluent Cloud bugs across two projects. The second time, I extracted a skill from my tapes sessions instead of debugging from scratch. Now every agent I run knows the fix.
Read more →Agents don't need to remember everything. They need to observe what went wrong. Structured logs become telemetry, telemetry becomes signals, and signals become the context for the next run.
Read more →Once agents got good enough to clone repos, install dependencies, modify files, and execute code they just wrote, the natural question became: where does this actually run? Sandboxes were designed for short-lived, contained execution. Agents aren't snippets.
Read more →An AI agent played Pokémon Red for 1,000 turns without leaving the bedroom. The failure wasn't the model. It was the absence of telemetry.
Read more →Every commercial aircraft carries a flight recorder. Agent systems don't. The gap between an agent that works and an agent you can trust in production is telemetry.
Read more →Optimizing for model performance while ignoring execution infrastructure is like choosing the fastest race car without checking whether the track has guardrails. The real engineering challenge isn't squeezing more capability out of LLMs—it's figuring out how to actually run them safely.
Read more →stereOS is a NixOS based operating system designed for AI agents to be secure, self-managed, and efficient
Read more →Agents are now consequential enough to warrant national-level oversight. Compliance doesn't require weaker AI—it requires observable AI. The observable era is the one that scales.
Read more →Visibility is the easiest piece. The hard part is analyzing and understanding what you're observing. The next phase of agent infrastructure won't be defined by visibility—it will be defined by durability.
Read more →If you're paying $100/month but consuming multiples of that in value, you have to start wondering when that's going to catch up to you. The AI coding tool market is following a familiar playbook.
Read more →Documentation that writes itself sounds like a gimmick. For tapes.dev, it's an operational reality. Here's how we wired continuous AI into our workflow to keep docs accurate as the codebase evolves.
Read more →How we solved the pain of cross-compiling Go projects with C dependencies by combining Zig as a drop-in compiler, Dagger for reproducible builds, and osxcross for macOS targets.
Read more →We built tapes to solve the observability gap in AI agent infrastructure. Every session recorded, every decision searchable, every conversation replayable.
Read more →Turn every session into knowledge at team scale.