10 posts tagged with "memory"

edit: yeah we built the thing in this post. still mad about the problem though.

support has had a memory-backed agent running for like three months. works. incidents retained. deploy region flips retained. recall in prod is actually correct. rare, I know.

then ML wanders over. "hey can we get JSONL for a fine-tune."

what goes in the zip file?

chat export. messages[]. maybe someone regex'd out the system prompt. no memory IDs. no "fact A got superseded on Feb 3." no link to the document chunk anyone actually trusted.

you fine-tune. model comes back confident. cites jack shit. confidently wrong about a deploy region you fixed in the bank six weeks ago.

you open the memory UI. bank was right. dataset was fiction.

that's not "the LLM hallucinated." that's two teams shipping different religions and nobody noticed until after the GPU invoice.

Atulya started as memory infrastructure for agents and teams. But a durable brain needs more than storage and retrieval. It also needs safe memory evolution. That is why Atulya now supports Git-like memory repos with branches, commits, rollback, and fork-to-bank workflows.

Storing memory easy. Pile of facts. Searchable. Done.

But pile of facts is not intelligence. Intelligence is knowing what matters in the pile. What is hot right now. What is fading. What just spiked abnormally. What contradicts something believed six months ago.

That is BRAIN. Layer inside Atulya that thinks about what got retained — continuously, in background, without waiting to be asked.

Agent forget. That core problem.

Not dramatic failure. Quiet one. Agent help Anurag refactor module today. Tomorrow — same agent, same repo — ask about that module. Agent say "I don't have context for that." Like nothing happened.

Worse: agent make change. Confident. No memory of constraint set three sessions ago. Break thing. No warning. Just breakage.

Agent remember fact. "Anurag reviewed PR #88." Good.

Agent remember another fact. "anurag@company.com approved deploy." Good.

Agent asked: "What did Anurag do this week?" Agent say: "I found one PR review. Nothing else."

Wrong. Two facts. Same person. Agent did not know. Looked like two separate entities. Completely unrelated.

This is entity fragmentation. Not edge case. Default state of any system that stores facts without resolving who they are about.

Build a ChatGPT-style loop with retain(), recall(), and reflect(). No vector DB, no embedding pipeline, no custom RAG.

Give any MCP client structured long-term memory: retain, recall, reflect, and mental models. One Docker command for local stack, or Atulya Cloud over HTTPS.

atulya-litellm intercepts LiteLLM completions: recall before the call, retain after. Works across OpenAI, Anthropic, Groq, Azure, Bedrock, Vertex, and 100+ providers with one bank.

CrewAI's built-in memory works inside one kickoff(). When the process exits, context is gone. atulya-crewai implements CrewAI's Storage interface on top of Atulya banks so knowledge compounds across runs.

TL;DR​

• Chat SDK bots start fresh every message unless you add memory
• @eight-atulya/atulya-chat wraps handlers with withAtulyaChat()
• One shared bank works across Slack, Discord, Teams, Google Chat
• Cookbook example tracks bugs and feature requests across platforms