my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

#1	---
#2	title: Advanced Memory Operations
#3	description: "Run richer add/search/update/delete flows on the managed platform with metadata, rerankers, and per-request controls."
#4	---
#5
#6	# Make Platform Memory Operations Smarter
#7
#8	<Info>
#9	Prerequisites
#10	- Platform workspace with API key
#11	- Python 3.10+ and Node.js 18+
#12	- Async memories enabled in your dashboard (Settings → Memory Options)
#13	</Info>
#14
#15	<Tip>
#16	Need a refresher on the core concepts first? Review the <Link href="/core-concepts/memory-operations/add">Add Memory</Link> overview, then come back for the advanced flow.
#17	</Tip>
#18
#19	## Install and authenticate
#20
#21	<Tabs>
#22	<Tab title="Python">
#23	<Steps>
#24	<Step title="Install the SDK with async extras">
#25	```bash
#26	pip install "mem0ai[async]"
#27	```
#28	</Step>
#29	<Step title="Export your API key">
#30	```bash
#31	export MEM0_API_KEY="sk-platform-..."
#32	```
#33	</Step>
#34	<Step title="Create an async client">
#35	```python
#36	import os
#37	from mem0 import AsyncMemoryClient
#38
#39	memory = AsyncMemoryClient(api_key=os.environ["MEM0_API_KEY"])
#40	```
#41	</Step>
#42	</Steps>
#43	</Tab>
#44	<Tab title="TypeScript">
#45	<Steps>
#46	<Step title="Install the OSS SDK">
#47	```bash
#48	npm install mem0ai
#49	```
#50	</Step>
#51	<Step title="Load your API key">
#52	```bash
#53	export MEM0_API_KEY="sk-platform-..."
#54	```
#55	</Step>
#56	<Step title="Instantiate the client">
#57	```typescript
#58	import { Memory } from "mem0ai";
#59
#60	const memory = new Memory({ apiKey: process.env.MEM0_API_KEY!, async: true });
#61	```
#62	</Step>
#63	</Steps>
#64	</Tab>
#65	</Tabs>
#66
#67	## Add memories with metadata and graph context
#68
#69	<Tabs>
#70	<Tab title="Python">
#71	<Steps>
#72	<Step title="Record conversations with metadata">
#73	```python
#74	conversation = [
#75	{"role": "user", "content": "I'm Morgan, planning a 3-week trip to Japan in May."},
#76	{"role": "assistant", "content": "Great! I'll track dietary notes and cities you mention."},
#77	{"role": "user", "content": "Please remember I avoid shellfish and prefer boutique hotels in Tokyo."},
#78	]
#79
#80	result = await memory.add(
#81	conversation,
#82	user_id="traveler-42",
#83	metadata={"trip": "japan-2025", "preferences": ["boutique", "no-shellfish"]},
#84	enable_graph=True,
#85	run_id="planning-call-1",
#86	)
#87	```
#88	</Step>
#89	</Steps>
#90	</Tab>
#91	<Tab title="TypeScript">
#92	<Steps>
#93	<Step title="Capture context-rich memories">
#94	```typescript
#95	const conversation = [
#96	{ role: "user", content: "I'm Morgan, planning a 3-week trip to Japan in May." },
#97	{ role: "assistant", content: "Great! I'll track dietary notes and cities you mention." },
#98	{ role: "user", content: "Please remember I avoid shellfish and love boutique hotels in Tokyo." },
#99	];
#100
#101	const result = await memory.add(conversation, {
#102	userId: "traveler-42",
#103	metadata: { trip: "japan-2025", preferences: ["boutique", "no-shellfish"] },
#104	enableGraph: true,
#105	runId: "planning-call-1",
#106	});
#107	```
#108	</Step>
#109	</Steps>
#110	</Tab>
#111	</Tabs>
#112
#113	<Info icon="check">
#114	Successful calls return memories tagged with the metadata you passed. In the dashboard, confirm a graph edge between “Morgan” and “Tokyo” and verify the `trip=japan-2025` tag exists.
#115	</Info>
#116
#117	## Retrieve and refine
#118
#119	<Tabs>
#120	<Tab title="Python">
#121	<Steps>
#122	<Step title="Filter by metadata + reranker">
#123	```python
#124	matches = await memory.search(
#125	"Any food alerts?",
#126	user_id="traveler-42",
#127	filters={"metadata.trip": "japan-2025"},
#128	rerank=True,
#129	include_vectors=False,
#130	)
#131	```
#132	</Step>
#133	<Step title="Update a memory inline">
#134	```python
#135	await memory.update(
#136	memory_id=matches["results"][0]["id"],
#137	content="Morgan avoids shellfish and prefers boutique hotels in central Tokyo.",
#138	)
#139	```
#140	</Step>
#141	</Steps>
#142	</Tab>
#143	<Tab title="TypeScript">
#144	<Steps>
#145	<Step title="Search with metadata filters">
#146	```typescript
#147	const matches = await memory.search("Any food alerts?", {
#148	userId: "traveler-42",
#149	filters: { "metadata.trip": "japan-2025" },
#150	rerank: true,
#151	includeVectors: false,
#152	});
#153	```
#154	</Step>
#155	<Step title="Apply an update">
#156	```typescript
#157	await memory.update(matches.results[0].id, {
#158	content: "Morgan avoids shellfish and prefers boutique hotels in central Tokyo.",
#159	});
#160	```
#161	</Step>
#162	</Steps>
#163	</Tab>
#164	</Tabs>
#165
#166	<Tip>
#167	Need to pause graph writes on a per-request basis? Pass `enableGraph: false` (TypeScript) or `enable_graph=False` (Python) when latency matters more than relationship building.
#168	</Tip>
#169
#170	## Clean up
#171
#172	<Tabs>
#173	<Tab title="Python">
#174	<Steps>
#175	<Step title="Delete scoped memories">
#176	```python
#177	await memory.delete_all(user_id="traveler-42", run_id="planning-call-1")
#178	```
#179	</Step>
#180	</Steps>
#181	</Tab>
#182	<Tab title="TypeScript">
#183	<Steps>
#184	<Step title="Remove the run">
#185	```typescript
#186	await memory.deleteAll({ userId: "traveler-42", runId: "planning-call-1" });
#187	```
#188	</Step>
#189	</Steps>
#190	</Tab>
#191	</Tabs>
#192
#193	## Quick recovery
#194
#195	- `Missing required key enableGraph`: update the SDK to `mem0ai>=0.4.0`.
#196	- `Graph backend unavailable`: retry with `enableGraph=False` and inspect your graph provider status.
#197	- Empty results with filters: log `filters` values and confirm metadata keys match (case-sensitive).
#198
#199	<Warning>
#200	Metadata keys become part of your filtering schema. Stick to lowercase snake_case (`trip_id`, `preferences`) to avoid collisions down the road.
#201	</Warning>
#202
#203	<CardGroup cols={2}>
#204	<Card
#205	title="Tune Metadata Filtering"
#206	description="Layer field-level filters on top of advanced operations."
#207	icon="funnel"
#208	href="/open-source/features/metadata-filtering"
#209	/>
#210	<Card
#211	title="Explore Reranker Search"
#212	description="See how rerankers boost accuracy after vector + graph retrieval."
#213	icon="sparkles"
#214	href="/open-source/features/reranker-search"
#215	/>
#216	</CardGroup>
#217

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public