Sklad, a key-value database in Zig
Lock-free skip-list memtable in Zig when Redis already dominates this space.
Agentic Task Queue - batch process tasks with agents in parallel.
Fresh context per task prevents bloat, but agent orchestration tools already exist.
Developers running batch LLM operations on large datasets
LangChain · CrewAI · DSPy
I had a huge number of database rows that each needed LLM reasoning and tool use.
Existing harnesses like Claude Code don't handle this well - give it a batch and it'll try to write you a script instead of actually reasoning over each item.
Even when it works, context bloats after each turn - gets expensive, slow, and starts hallucinating. And there's no built-in retry handling or parallelization.
So I built this tool. Define the task as a prompt, pipe in your task list, set concurrency, model, tools, etc.
Open to feedback!
Lock-free skip-list memtable in Zig when Redis already dominates this space.
Task queue for AI agents, but orchestrates existing tools without novel architecture.
push.rocks/taskbuffer appears to be a focused utility for queuing and scheduling short-lived tasks — useful when you need to debounce bursts or smooth calls to a rate-limited API. The npm landing (blocked by Cloudflare in my screenshot) doesn't show whether it offers persistence, retries, or clustering, so right now it looks like a handy single-process tool but not a replacement for Bull/Agenda in production workloads.
Behavior-tree orchestration for agents when LangGraph and AutoGen already exist.
CBT-grounded anxiety container, but meditation app category already flooded.
On-device AI for work inbox, but landing page makes promises without shipped product.