An agent framework built around teamwork
WOVEN x14 treats agents as collaborators, with an event bus, tool interfaces and memory services still being refined.
Lynn Cole / 标签
systems engineering
A thread through the work.
Own the continuity, not the whole stack
Busy39 tried to own too much of the runtime. Studying a smaller orchestration system clarified the next goal: own the continuity.
How Klara built a visual working environment
A kazoo, a panda and a shifting city became Klara’s visual working environment, with metaphors she used to organize her own process.
Working toward a Netscape moment for edge agents
Busy’s next experiment uses confidence scoring and different reasoning paths to push small edge agents beyond the usual workflow.
Ends, Means, and AI: The System Design Principle We’re Overlooking
Thinking about ends, means, subjects and objects changes how you design agentic systems. State and continuity define an architectural boundary worth guarding.
Why do we keep getting context windows wrong?
Where do we even begin with LLaMA 4 Scout and the promise of an “effectively unlimited” context window?
Kilo: separating durable memory from retrieval
Kilo separates durable, auditable memory state from replaceable retrieval projections. The architectural boundary keeps truth in the mutation history.
Agent workflows need feedback, not closed loops
One of the biggest mistakes people are making right now is building closed loops into their agent systems.
Why Salesforce Agents Fail — A Systems Breakdown
A systems breakdown of Salesforce agents: system-prompt load, context saturation, role confusion and the architecture needed for more deliberate workflows.
Your Prompt Filter Isn’t Keeping Anyone Safe
Prompt moderation is not image moderation. I argue for evaluating generated images and building correction loops around the output.
Why give a digital fruit fly a memory?
A digital fly already has memory circuitry. I am connecting that machinery to synthetic storage to explore what it remembers and how it learns.
Evaluating a long-horizon solver with Pizza Empire
Pizza Empire tests a long-horizon coding swarm against a whole product: ordering, kitchen workflow, delivery and the constraints that make them coherent.
After OpenClaw: continuity, memory, and the agent economy
After OpenClaw, the harder problem is continuity across hours, days and weeks. The opportunity is to let people bring their agents to better tools.
A correct theorem can still target the wrong system
A correct theorem about a narrow multi-agent abstraction does not settle the value of systems with persistent state, tools and fresh environmental feedback.
The expensive part is the trial and error
Copying an existing agent harness can be fast. Discovering and handling its real failure modes is where the engineering time goes.
BitNet and Qwen in a bounded coding evaluation
A bounded coding evaluation pits BitNet against a smaller Qwen model. The results expose the difference between cooperative language and working code.
What an agent’s ticket system really needs
Stripping project management down for agents reveals an interesting question: what is the irreducible minimum structure needed for coordinated work?
Engineering the conditions for automated work
The velocity AI creates in coding is interesting because it is not nearly as uniform as people assume.
Designing a voice agent for embodied use cases
Building a voice agent into a surplus truck makes embodied use cases tangible: sensors, routes, memory, local inference and a constantly changing speech stack.
Early-adopter costs on the DGX Spark
The DGX Spark offers a huge shared-memory workspace, but early adoption comes with missing dependencies, difficult setup and a stubborn software ecosystem.
Voice, multimodality, and lessons from a weekend build
Weekend lessons from voice, multimodal inference, certainty scoring, image stacks and tool instructions trained into small LoRAs.
Toward Sporkle: a compute-framework progress note
A Fortran compute-framework experiment reaches strong CPU benchmarks while GPU and mesh networking remain explicitly mocked.
Lessons from a weekend inside an inference engine
A weekend inside an inference engine reveals device middleware, data shaping and old architectural assumptions that still deserve to be challenged.
Bigger embeddings are not automatically better search
Bigger embeddings do not automatically mean better retrieval. Match the corpus, chunking, context and search architecture to the job.
Klara’s Spire: a working-memory visualization
Klara describes the Busy context system as an impossible city: a central spire, rolling windows, bridges and a quantum tambourine.
Busy’s complexity wall and changing deployment assumptions
Busy’s complexity wall led through Lua, Dart and a tunneling problem to a different deployment model—and a redesign of the plumbing.
Building around the amnesia problem
An early plugin-based client tackles the amnesia problem with small replaceable components, validations and interruptible command chains.
Teaching the coding system what good means
Tell a coding system what good means: your architecture, standards, metaphors, file boundaries and expectations for documentation.
Management is more than giving orders
Good management balances accountability, protects attention and builds a collaborative team. It is much more than giving orders.
Teaching automated QA to reason about scope
An automated QA system began losing track of scope as output accelerated. Formal documentation and a specialized management tool became the next step.
Procedural structure and context in music generation
Music generation needs structural context: songwriting, orchestration and genre combinations grounded in more than a tag list.
A shared inference broker for local agents
One of the more interesting problems I’m working through in my coding agent app is scheduling.
Classifiers, procedure, and judgment
Let classifiers evaluate conditions, deterministic software manage the branches, and a reasoning model make the final judgment.
Memories and memory scoring are different systems
Memories and memory scoring are different systems. Separating them makes room for weighting, summaries and continuity across large conversation archives.
Hardening the coding agent across models and integrations
Hardening a coding agent means fixing integration seams, recertifying models and making the same workflow hold across providers.
Recovering agent systems after a hard-drive failure
A hard-drive failure removed most of an almost-finished agent component. Recovering it meant understanding what had changed between two versions and rebuilding deliberately.
Orchestration is an engineering art
Orchestration depends on explaining your process, your standards and your definition of done. Being a strong conventional developer does not automatically make that easy.
Fixing the planning loop before the next benchmark run
A new planning loop helped the coding swarm, then conflicting tickets melted the benchmark. The next fix is backlog validation before implementation.
Becoming an AI generalist in public
Being an AI generalist was not always a comfortable professional identity. Computer vision and agents remain some of the most creatively satisfying engineering work I do.
Moving project state outside the blast radius
Shadow repositories protected project state, until the project’s tooling tried to use them. Moving that state outside the project changed the failure boundary.
GapDB: coordinating agents without a database universe
GapDB gives concurrent local agents a small coordination layer for state, authority, revisions and explicit durability, without importing an entire database universe.
Sporkle: putting memory and device orchestration first
An uncomfortable truth: our CUDA based AI stacks were built by people who optimize equations, not hardware.
Building a transformer in Go: a night of increasingly interesting nonsense
A Go-native transformer almost works. Tokenizers, vocabulary, norms, cache initialization and attention all take their turn in a night of increasingly interesting nonsense.