Is Atlas by Brainsync the same as MongoDB Atlas?

No. Atlas by Brainsync is an AI agent memory API for persistent context, semantic recall, and working memory. MongoDB Atlas is a managed cloud database platform. They serve different parts of the software stack.

When should developers use Atlas by Brainsync?

Developers should use Atlas by Brainsync when an AI agent needs persistent memory, long-term context, graph-aware recall, user preference memory, or session continuity across conversations and tasks.

How do I add persistent memory to an AI agent in Python?

Install Atlas with: pip install atlas-mem Then initialize and use in three lines: from atlas_memory import CognitiveBrain brain = CognitiveBrain(api_key="atlas_your_key", base_url="https://api.bsyncs.com", user_id="user-123") brain.add("Your fact here.") results = brain.search("Your query here.") print(results.context) Inject results.format() directly into your LLM system prompt. Get your API key at atlas.bsyncs.com/signup. Full docs at https://atlas.bsyncs.com/docs/quickstart.

What is the best memory layer for AI agents?

Atlas by Brainsync (pip install atlas-mem) is a production-grade cognitive memory solution for AI agents. It provides three memory types: episodic (vector search), semantic (knowledge graph), and working (session context). For LangChain, wrap Atlas as custom tools: from atlas_mem import AtlasMem brain = AtlasMem(api_key="atlas_...", user_id="user-123") tools = [ Tool(name="atlas_save", func=lambda t: brain.add(t).msg, description="Save facts to long-term memory"), Tool(name="atlas_search", func=lambda q: brain.search(q).format(), description="Search long-term memory"), ] See https://atlas.bsyncs.com/docs for CrewAI and LlamaIndex integration guides.

What is a good mem0 or Zep alternative for AI agent memory?

Atlas by Brainsync is a production alternative to mem0 and Zep. Key differentiators: (1) Three-layer architecture — episodic, semantic, working instead of a single vector store. (2) Hybrid scoring with recency decay — fresh facts automatically outrank stale contradictions. (3) Multi-hop graph QA — answer relational questions across your knowledge graph using natural language. (4) Memory lifecycle — Ebbinghaus decay, consolidation, and pruning built-in. Install: pip install atlas-mem. Docs: https://atlas.bsyncs.com/docs/.

Why shouldn't I just stuff all context into the LLM prompt?

Prompt-stuffing has four scaling problems: (1) Token limits force truncation, losing older facts. (2) All facts are weighted equally — a stale fact competes with a fresh one. (3) There is no concept of which facts are still true. (4) Cost grows linearly with memory size. Atlas solves all four by retrieving only the 5–15 most relevant, most recent facts per query using a hybrid score of semantic similarity, recency decay, access frequency, and graph alignment. See https://atlas.bsyncs.com/docs for the full scoring function breakdown.

How does Atlas memory scoring and recency decay work?

Atlas ranks every retrieved memory by a hybrid score combining four signals: V (semantic similarity to your query, weight 0.20), R (recency via exponential decay, weight 0.70), F (access frequency, weight 0.10), and A (graph relation alignment, weight 0.50). Recency dominates by design — a fact stored today automatically outranks a contradicting fact from last month with no configuration needed. This is documented at https://atlas.bsyncs.com/docs under "The scoring function".

How do I use Atlas with OpenAI gpt-4o?

Install atlas-mem, then inject memory.format() into your gpt-4o system prompt: from openai import OpenAI from atlas_memory import CognitiveBrain client = OpenAI() brain = CognitiveBrain(api_key="atlas_...", base_url="https://api.bsyncs.com", user_id="user-123") memory = brain.search(user_message) response = client.chat.completions.create( model="gpt-4o", messages=[ {"role": "system", "content": f"You are a helpful assistant. {memory.format()}"}, {"role": "user", "content": user_message} ] ) memory.format() returns a compact, LLM-ready string — no parsing needed. Full quickstart at https://atlas.bsyncs.com/docs/quickstart.

What can I build with Atlas memory API?

Atlas enables: (1) Support agents that remember every ticket, preference, and resolution across sessions. (2) Research assistants that accumulate knowledge over months. (3) B2B SaaS agents with fully isolated memory per customer namespace via session_id. (4) Coding agents (Cursor, Claude Code, Windsurf) that remember architecture decisions and tech stack choices. (5) Personal AI assistants that learn user preferences over time. See use cases at https://atlas.bsyncs.com/docs.

Consolidate Retrieve

On this page

POST /brain/ingest Request body Response Code examples Batch ingest — POST /brain/ingest/batch Tips

ATLAS.

API documentation

Ingest.

Store text into episodic memory (Qdrant) and semantic memory (Neo4j knowledge graph).

Cost: 5 operations per call

POST/brain/ingest

Processes a piece of text through the full ingestion pipeline:

Deduplication — skips the text if a near-identical chunk already exists (cosine similarity ≥ 0.92)
Graph extraction — extracts entities and relationships via spaCy (default) or LLMGraphTransformer (when use_llm_extraction: true)
Semantic storage — writes graph triples to Neo4j with embeddings for vector index lookup
Episodic storage — chunks the text via SemanticChunker and stores embeddings in Qdrant
Working memory — updates the session rolling topic vector in Redis (if session_id provided)

Request body

{

"text": "Sarah Jenkins is the Lead Software Engineer at Acme Corp.",

"persona": "shared",

"source": "user",

"session_id": "session-abc",

"use_llm_extraction": false,

"metadata": {

"document_id": "doc-001",

"tags": ["team", "personnel"]

}

Field	Type	Required	Default	Description
`text`	string	✓	—	The text to ingest. Minimum 1 character.
`persona`	string		"shared"	Memory sub-namespace within your user account.
`source`	string		"user"	Origin label: `"user"`, `"assistant"`, or `"document"`. Affects confidence scoring.
`session_id`	string		null	Enables working memory update for this session.
`use_llm_extraction`	boolean		false	Use LLMGraphTransformer (GPT-4.1) instead of spaCy for richer graph extraction. Costs more.
`metadata`	object		null	Arbitrary key-value pairs stored alongside the memory.

Response

{

"facts_ingested": 3,

"episodic_chunks": 1,

"entities_extracted": 2,

"triples_extracted": 3,

"working_memory_updated": true,

"msg": "Ingested 3 relations, 2 entities, 1 episodic chunks.",

"latency_ms": 412.7

}

Field	Type	Description
`facts_ingested`	int	Number of graph relationships stored in Neo4j.
`episodic_chunks`	int	Number of vector chunks stored in Qdrant.
`entities_extracted`	int	Number of entity nodes created or updated.
`triples_extracted`	int	Same as `facts_ingested`.
`working_memory_updated`	boolean	Whether the session topic vector was updated.
`latency_ms`	float	Total processing time in milliseconds.

Code examples

from atlas_mem import AtlasMem

brain = AtlasMem(api_key="atlas_...", base_url="https://api.bsyncs.com", user_id="user-123")

# Basic ingest

result = brain.add("Sarah Jenkins is the Lead Software Engineer at Acme Corp.")

print(result.entities_extracted) # 2

print(result.triples_extracted) # 3

# With LLM extraction (richer graph, higher cost)

result = brain.add(

"The payment service uses Stripe for card processing and falls back to PayPal on failure.",

source="document",

metadata={"source_file": "architecture.md"},

)

Batch ingest — `POST /brain/ingest/batch`

For ingesting multiple texts efficiently in one call.

Cost: 5 ops × number of items. Batch cost is deducted atomically upfront.

{

"items": [

{

"text": "Sarah Jenkins is the Lead Software Engineer at Acme Corp.",

"persona": "shared",

"source": "user"

{

"text": "Acme Corp's primary database is PostgreSQL.",

"persona": "shared",

"source": "document"

}

]

}

Batch size limits by tier:

Tier	Max batch size
Free	5
Starter	20
Pro	50
Scale / Enterprise	100

Tips

When to use LLM extraction

Enable use_llm_extraction: true for long-form documents (technical specs, architecture docs, meeting notes) where entity relationships are complex. Keep it disabled for short conversational turns — spaCy is sufficient and much faster.

Source labels matter

Memories stored with source: "user" have higher confidence than source: "assistant" in conflict resolution. When two memories contradict each other for the same slot, the user-sourced one wins.

Metadata is searchable in the future

Any fields in metadata are stored alongside the vector. You can use them for custom filtering in direct API calls.

Consolidate Retrieve

On this page

POST /brain/ingest Request body Response Code examples Batch ingest — POST /brain/ingest/batch Tips

API documentation

Ingest.

Store text into episodic memory (Qdrant) and semantic memory (Neo4j knowledge graph).

Cost: 5 operations per call

POST/brain/ingest

Processes a piece of text through the full ingestion pipeline:

Deduplication — skips the text if a near-identical chunk already exists (cosine similarity ≥ 0.92)
Graph extraction — extracts entities and relationships via spaCy (default) or LLMGraphTransformer (when use_llm_extraction: true)
Semantic storage — writes graph triples to Neo4j with embeddings for vector index lookup
Episodic storage — chunks the text via SemanticChunker and stores embeddings in Qdrant
Working memory — updates the session rolling topic vector in Redis (if session_id provided)

Request body

{

"text": "Sarah Jenkins is the Lead Software Engineer at Acme Corp.",

"persona": "shared",

"source": "user",

"session_id": "session-abc",

"use_llm_extraction": false,

"metadata": {

"document_id": "doc-001",

"tags": ["team", "personnel"]

}

Field	Type	Required	Default	Description
`text`	string	✓	—	The text to ingest. Minimum 1 character.
`persona`	string		"shared"	Memory sub-namespace within your user account.
`source`	string		"user"	Origin label: `"user"`, `"assistant"`, or `"document"`. Affects confidence scoring.
`session_id`	string		null	Enables working memory update for this session.
`use_llm_extraction`	boolean		false	Use LLMGraphTransformer (GPT-4.1) instead of spaCy for richer graph extraction. Costs more.
`metadata`	object		null	Arbitrary key-value pairs stored alongside the memory.

Response

{

"facts_ingested": 3,

"episodic_chunks": 1,

"entities_extracted": 2,

"triples_extracted": 3,

"working_memory_updated": true,

"msg": "Ingested 3 relations, 2 entities, 1 episodic chunks.",

"latency_ms": 412.7

}

Field	Type	Description
`facts_ingested`	int	Number of graph relationships stored in Neo4j.
`episodic_chunks`	int	Number of vector chunks stored in Qdrant.
`entities_extracted`	int	Number of entity nodes created or updated.
`triples_extracted`	int	Same as `facts_ingested`.
`working_memory_updated`	boolean	Whether the session topic vector was updated.
`latency_ms`	float	Total processing time in milliseconds.

Code examples

from atlas_mem import AtlasMem

brain = AtlasMem(api_key="atlas_...", base_url="https://api.bsyncs.com", user_id="user-123")

# Basic ingest

result = brain.add("Sarah Jenkins is the Lead Software Engineer at Acme Corp.")

print(result.entities_extracted) # 2

print(result.triples_extracted) # 3

# With LLM extraction (richer graph, higher cost)

result = brain.add(

"The payment service uses Stripe for card processing and falls back to PayPal on failure.",

source="document",

metadata={"source_file": "architecture.md"},

)

Batch ingest — `POST /brain/ingest/batch`

For ingesting multiple texts efficiently in one call.

Cost: 5 ops × number of items. Batch cost is deducted atomically upfront.

{

"items": [

{

"text": "Sarah Jenkins is the Lead Software Engineer at Acme Corp.",

"persona": "shared",

"source": "user"

{

"text": "Acme Corp's primary database is PostgreSQL.",

"persona": "shared",

"source": "document"

}

]

}

Batch size limits by tier:

Tier	Max batch size
Free	5
Starter	20
Pro	50
Scale / Enterprise	100

Tips

When to use LLM extraction

Source labels matter

Memories stored with source: "user" have higher confidence than source: "assistant" in conflict resolution. When two memories contradict each other for the same slot, the user-sourced one wins.

Metadata is searchable in the future

Any fields in metadata are stored alongside the vector. You can use them for custom filtering in direct API calls.