For: Qwen Cloud Global AI Hackathon 2026 · Track: MemoryAgent · Deadline: July 20, 2026, 2:00pm PDT Prize target: $7,000 cash + $3,000 cloud credits (track win)
AI agents forget everything between sessions. Every conversation starts from zero. The fix most people build is "dump all past chat into a vector DB and stuff it into the prompt" — which doesn't scale, never forgets noise, and isn't reusable across agents.
An MCP server that gives ANY agent a self-curating long-term memory. It:
- Decides what's worth remembering — a Qwen3 agent extracts durable facts/preferences/decisions from each interaction (ignores small talk).
- Stores with a salience + decay score — embeddings + a forgetting curve, so stale one-off details fade while important facts persist.
- Retrieves only what matters — semantic similarity × recency-decay × salience, top-k into the prompt.
- Is reusable infrastructure — because it's an MCP server, the same memory plugs into a chat assistant, a coding assistant, anything.
Agent memory is crowded (Mem0, Zep, Letta/MemGPT). Cortex does NOT compete on "we store memories." Its wedge is what incumbents under-serve:
- Active, salience-scaled forgetting — context stays lean and self-prunes; stale memories actually leave (most systems only grow). Proven by
eval.py(89% footprint shrink, 0 important lost). - Transparency & governance — memory you can see, audit, edit, and delete (the live memory-graph UI). The open gap for enterprise/regulated + consumer-trust use.
- MCP-native — a drop-in memory primitive any agent inherits, no SDK lock-in.
Positioning line: transparent, self-pruning memory — the agent memory you can audit, and that doesn't bloat.
Across 3 sessions days apart, a generic assistant wired to Cortex:
- S1: learns the user is vegan + nut-allergic + prefers terse answers.
- S2: proactively avoids a recommendation that conflicts.
- S3: weeks of "noise" later, still recalls the allergy but has forgotten a throwaway detail.
- Live memory-graph panel visualizes writes / retrievals / scores / forgets in real time.
- Closer: swap Cortex onto a different agent (a coding assistant) — the memory carries over. Proof it's infra, not an app.
- Innovation 30% — it is an MCP integration + the remember/forget policy is a custom skill (the two things the rubric names by hand).
- Tech Depth 30% — standalone MCP server + memory engine (extract → score → decay → rerank), modular, reusable.
- Impact 25% — universal pain; shipped as open-source MCP infra → maxes "scalable to community."
- Presentation 15% — the memory-graph panel literally "visualizes the key logic" (their phrase); cross-agent swap is the money shot.
- Backend (Python / FastAPI):
cortex/package —memory_engine.py(extract/score/decay/retrieve),qwen_client.py(DashScope OpenAI-compatible),mcp_server.py(FastMCP:remember,recalltools),main.py(multi-tenant REST API:/remember/recall/chat/memories/audit). - Frontend:
web/index.html— a single-page, dependency-free HTML/CSS/JS live memory-graph UI (writes/retrievals/decay/forgets rendered in real time, multi-agent swap, audit drawer, memory inspector). Shipped; no separate Next.js app. - Store: shared persistent JSON store (
store.py) so the MCP server + REST clients + multiple agents see the same memories (what makes the cross-agent swap real, not simulated). - Deploy: Alibaba Cloud (mandatory pass/fail gate — deploy early, capture proof link).
- D1–2: DashScope key + $40 voucher; port LLM calls to Qwen (OpenAI-compatible); deploy a hello-world to Alibaba Cloud to prove the gate early.
- D3–5: memory engine — extraction, embeddings, salience+decay, retrieve/rerank, forget. ✅ (skeleton in repo)
- D6–7: MCP server exposing
remember/recall; shared persistent store. - D8–10:
web/index.htmlchat client + live memory-graph panel. ✅ (shipped) - D11–12: wire a 2nd agent (coding assistant) to the same Cortex for the swap demo.
- D13–14: scripted 3-session demo data + polish UX.
- D15: deploy final to Alibaba Cloud; README + architecture diagram.
- D16: record ≤3-min demo (lead with cross-agent swap).
- D17: Devpost submission (video + repo + slides + description) buffer.
- Pre-record the cross-agent swap instead of live.
- Memory-graph → simpler real-time "memory ledger" table (still "visualizes key logic").
- Cut S2 proactive-avoidance; keep S3 forget-vs-recall (the non-negotiable core).
MCP server + Qwen-driven remember/forget engine + one client showing recall-after-noise and forgetting, deployed on Alibaba Cloud, with a logic-visualizing demo.