Metadata-Version: 2.4
Name: remem-ai
Version: 0.6.0
Summary: An AI Work Reuse Engine — remember expensive AI work and reuse it intelligently.
Author: Harshvardhan Singh
License:                                  Apache License
                                   Version 2.0, January 2004
                                http://www.apache.org/licenses/
        
           TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
        
           1. Definitions.
        
              "License" shall mean the terms and conditions for use, reproduction,
              and distribution as defined by Sections 1 through 9 of this document.
        
              "Licensor" shall mean the copyright owner or entity authorized by
              the copyright owner that is granting the License.
        
              "Legal Entity" shall mean the union of the acting entity and all
              other entities that control, are controlled by, or are under common
              control with that entity. For the purposes of this definition,
              "control" means (i) the power, direct or indirect, to cause the
              direction or management of such entity, whether by contract or
              otherwise, or (ii) ownership of fifty percent (50%) or more of the
              outstanding shares, or (iii) beneficial ownership of such entity.
        
              "You" (or "Your") shall mean an individual or Legal Entity
              exercising permissions granted by this License.
        
              "Source" form shall mean the preferred form for making modifications,
              including but not limited to software source code, documentation
              source, and configuration files.
        
              "Object" form shall mean any form resulting from mechanical
              transformation or translation of a Source form, including but
              not limited to compiled object code, generated documentation,
              and conversions to other media types.
        
              "Work" shall mean the work of authorship, whether in Source or
              Object form, made available under the License, as indicated by a
              copyright notice that is included in or attached to the work
              (an example is provided in the Appendix below).
        
              "Derivative Works" shall mean any work, whether in Source or Object
              form, that is based on (or derived from) the Work and for which the
              editorial revisions, annotations, elaborations, or other modifications
              represent, as a whole, an original work of authorship. For the purposes
              of this License, Derivative Works shall not include works that remain
              separable from, or merely link (or bind by name) to the interfaces of,
              the Work and Derivative Works thereof.
        
              "Contribution" shall mean any work of authorship, including
              the original version of the Work and any modifications or additions
              to that Work or Derivative Works thereof, that is intentionally
              submitted to Licensor for inclusion in the Work by the copyright owner
              or by an individual or Legal Entity authorized to submit on behalf of
              the copyright owner. For the purposes of this definition, "submitted"
              means any form of electronic, verbal, or written communication sent
              to the Licensor or its representatives, including but not limited to
              communication on electronic mailing lists, source code control systems,
              and issue tracking systems that are managed by, or on behalf of, the
              Licensor for the purpose of discussing and improving the Work, but
              excluding communication that is conspicuously marked or otherwise
              designated in writing by the copyright owner as "Not a Contribution."
        
              "Contributor" shall mean Licensor and any individual or Legal Entity
              on behalf of whom a Contribution has been received by Licensor and
              subsequently incorporated within the Work.
        
           2. Grant of Copyright License. Subject to the terms and conditions of
              this License, each Contributor hereby grants to You a perpetual,
              worldwide, non-exclusive, no-charge, royalty-free, irrevocable
              copyright license to reproduce, prepare Derivative Works of,
              publicly display, publicly perform, sublicense, and distribute the
              Work and such Derivative Works in Source or Object form.
        
           3. Grant of Patent License. Subject to the terms and conditions of
              this License, each Contributor hereby grants to You a perpetual,
              worldwide, non-exclusive, no-charge, royalty-free, irrevocable
              (except as stated in this section) patent license to make, have made,
              use, offer to sell, sell, import, and otherwise transfer the Work,
              where such license applies only to those patent claims licensable
              by such Contributor that are necessarily infringed by their
              Contribution(s) alone or by combination of their Contribution(s)
              with the Work to which such Contribution(s) was submitted. If You
              institute patent litigation against any entity (including a
              cross-claim or counterclaim in a lawsuit) alleging that the Work
              or a Contribution incorporated within the Work constitutes direct
              or contributory patent infringement, then any patent licenses
              granted to You under this License for that Work shall terminate
              as of the date such litigation is filed.
        
           4. Redistribution. You may reproduce and distribute copies of the
              Work or Derivative Works thereof in any medium, with or without
              modifications, and in Source or Object form, provided that You
              meet the following conditions:
        
              (a) You must give any other recipients of the Work or
                  Derivative Works a copy of this License; and
        
              (b) You must cause any modified files to carry prominent notices
                  stating that You changed the files; and
        
              (c) You must retain, in the Source form of any Derivative Works
                  that You distribute, all copyright, patent, trademark, and
                  attribution notices from the Source form of the Work,
                  excluding those notices that do not pertain to any part of
                  the Derivative Works; and
        
              (d) If the Work includes a "NOTICE" text file as part of its
                  distribution, then any Derivative Works that You distribute must
                  include a readable copy of the attribution notices contained
                  within such NOTICE file, excluding those notices that do not
                  pertain to any part of the Derivative Works, in at least one
                  of the following places: within a NOTICE text file distributed
                  as part of the Derivative Works; within the Source form or
                  documentation, if provided along with the Derivative Works; or,
                  within a display generated by the Derivative Works, if and
                  wherever such third-party notices normally appear. The contents
                  of the NOTICE file are for informational purposes only and
                  do not modify the License. You may add Your own attribution
                  notices within Derivative Works that You distribute, alongside
                  or as an addendum to the NOTICE text from the Work, provided
                  that such additional attribution notices cannot be construed
                  as modifying the License.
        
              You may add Your own copyright statement to Your modifications and
              may provide additional or different license terms and conditions
              for use, reproduction, or distribution of Your modifications, or
              for any such Derivative Works as a whole, provided Your use,
              reproduction, and distribution of the Work otherwise complies with
              the conditions stated in this License.
        
           5. Submission of Contributions. Unless You explicitly state otherwise,
              any Contribution intentionally submitted for inclusion in the Work
              by You to the Licensor shall be under the terms and conditions of
              this License, without any additional terms or conditions.
              Notwithstanding the above, nothing herein shall supersede or modify
              the terms of any separate license agreement you may have executed
              with Licensor regarding such Contributions.
        
           6. Trademarks. This License does not grant permission to use the trade
              names, trademarks, service marks, or product names of the Licensor,
              except as required for reasonable and customary use in describing the
              origin of the Work and reproducing the content of the NOTICE file.
        
           7. Disclaimer of Warranty. Unless required by applicable law or
              agreed to in writing, Licensor provides the Work (and each
              Contributor provides its Contributions) on an "AS IS" BASIS,
              WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
              implied, including, without limitation, any warranties or conditions
              of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
              PARTICULAR PURPOSE. You are solely responsible for determining the
              appropriateness of using or redistributing the Work and assume any
              risks associated with Your exercise of permissions under this License.
        
           8. Limitation of Liability. In no event and under no legal theory,
              whether in tort (including negligence), contract, or otherwise,
              unless required by applicable law (such as deliberate and grossly
              negligent acts) or agreed to in writing, shall any Contributor be
              liable to You for damages, including any direct, indirect, special,
              incidental, or consequential damages of any character arising as a
              result of this License or out of the use or inability to use the
              Work (including but not limited to damages for loss of goodwill,
              work stoppage, computer failure or malfunction, or any and all
              other commercial damages or losses), even if such Contributor
              has been advised of the possibility of such damages.
        
           9. Accepting Warranty or Additional Liability. While redistributing
              the Work or Derivative Works thereof, You may choose to offer,
              and charge a fee for, acceptance of support, warranty, indemnity,
              or other liability obligations and/or rights consistent with this
              License. However, in accepting such obligations, You may act only
              on Your own behalf and on Your sole responsibility, not on behalf
              of any other Contributor, and only if You agree to indemnify,
              defend, and hold each Contributor harmless for any liability
              incurred by, or claims asserted against, such Contributor by reason
              of your accepting any such warranty or additional liability.
        
           END OF TERMS AND CONDITIONS
        
           APPENDIX: How to apply the Apache License to your work.
        
              To apply the Apache License to your work, attach the following
              boilerplate notice, with the fields enclosed by brackets "[]"
              replaced with your own identifying information. (Don't include
              the brackets!)  The text should be enclosed in the appropriate
              comment syntax for the file format. We also recommend that a
              file or class name and description of purpose be included on the
              same "printed page" as the copyright notice for easier
              identification within third-party archives.
        
           Copyright 2024 Harshvardhan Singh
        
           Licensed under the Apache License, Version 2.0 (the "License");
           you may not use this file except in compliance with the License.
           You may obtain a copy of the License at
        
               http://www.apache.org/licenses/LICENSE-2.0
        
           Unless required by applicable law or agreed to in writing, software
           distributed under the License is distributed on an "AS IS" BASIS,
           WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
           See the License for the specific language governing permissions and
           limitations under the License.
        
Project-URL: Homepage, https://github.com/harshvardhansingh7/remem
Project-URL: Repository, https://github.com/harshvardhansingh7/remem
Project-URL: Issues, https://github.com/harshvardhansingh7/remem/issues
Keywords: ai,llm,rag,cache,semantic-cache,embeddings,reuse,agents,vector,infrastructure
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Typing :: Typed
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.21
Provides-Extra: dev
Requires-Dist: pytest>=7.0; extra == "dev"
Requires-Dist: ruff>=0.1; extra == "dev"
Dynamic: license-file

# Remem

> **Remember expensive AI work. Reuse it intelligently.**

<p align="center">
  <strong>An AI Work Reuse Engine for Retrieval-Augmented Generation (RAG), AI Agents, and LLM Applications.</strong>
</p>

<p align="center">
  <img alt="Python" src="https://img.shields.io/badge/Python-3.10+-blue.svg">
  <img alt="Status" src="https://img.shields.io/badge/Status-v0.6.0--alpha-orange">
  <img alt="License" src="https://img.shields.io/badge/License-Apache%202.0-green.svg">
</p>

---

# Installation

```bash
pip install remem
```

Remem has a single runtime dependency (`numpy`) and works on Python 3.10+.

## Quickstart

Integrating Remem takes a few lines. Wrap any expensive AI step in
`get_or_compute`: pass the query embedding and a callback that does the real
work. Remem decides whether to reuse a previous execution or run the callback,
then remembers the result for next time.

```python
from remem import Client, ExecutionResult

client = Client()  # sensible defaults; durable JSON persistence out of the box

def expensive_pipeline():
    docs = search_vector_db(query)        # your retrieval
    answer = call_llm(query, docs)        # your LLM call
    return ExecutionResult(response=answer, references=docs)

outcome = client.get_or_compute(
    query_embedding=embed(query),         # your embedding model
    compute_callback=expensive_pipeline,
)

print(outcome.decision)   # RESPONSE_REUSED | RETRIEVAL_USED | MISS
print(outcome.result)     # the answer (reused or freshly computed)
```

Need ephemeral storage (tests, notebooks)? Swap the backend without touching
anything else:

```python
from remem import Client, InMemoryStorage

client = Client(storage_backend=InMemoryStorage())  # nothing written to disk
```

Gradually adopt the advanced features when you need them — execution contexts,
custom policies, and metadata matching:

```python
from remem import Client, ExecutionContext, ReusePolicy

client = Client(
    policy=ReusePolicy(
        retrieval_threshold=0.80,   # reuse retrieved docs above this similarity
        response_threshold=0.95,    # reuse the whole LLM response above this
        require_same_model=True,    # never reuse across different models
    )
)

context = ExecutionContext(namespace="hr-bot", kb_version="2024.1", model="gpt-4o")
outcome = client.get_or_compute(embed(query), expensive_pipeline, context=context)
```

A complete, runnable RAG example lives in [`examples/rag_reuse.py`](examples/rag_reuse.py).

---

# Why Remem?

Modern AI applications repeatedly execute expensive operations for requests that are often semantically similar.

A typical AI workflow may involve:

* Generating embeddings
* Searching vector databases
* Retrieving knowledge chunks
* Reranking retrieved documents
* Constructing prompts
* Calling Large Language Models (LLMs)
* Executing tools or SQL queries
* Running multi-step agent workflows

Although user queries may be phrased differently, they frequently require nearly identical computation.

For example:

```text
"What is our company's vacation policy?"

"What are the PTO rules?"

"How many paid leaves do employees receive?"
```

Most AI systems execute the entire pipeline independently for each request, even when much of the previous work could be safely reused.

This leads to:

* Higher inference costs
* Increased latency
* Unnecessary vector searches
* Repeated reranking
* Duplicate LLM calls
* Repeated tool execution
* Wasted infrastructure resources

Remem exists to eliminate redundant AI work.

Instead of treating every request as completely new, Remem remembers previously executed work and determines whether any part of it can be safely reused.

---

# Vision

Remem is an **AI Work Reuse Engine**.

Rather than replacing your existing infrastructure (Redis, Postgres, Pinecone, Qdrant, Weaviate, Milvus, pgvector, or custom retrieval systems), Remem integrates alongside your application and continuously observes expensive AI operations.

When a new request arrives, Remem analyzes previously completed work and determines the highest level of computation that can be safely reused.

Depending on the request, this may include:

* Previously generated LLM responses
* Retrieved knowledge chunks
* Reranked search results
* Tool outputs
* SQL query results
* Agent execution artifacts
* Other reusable intermediate computations

If no reusable work exists, the application executes normally, and Remem learns from the new execution for future requests.

The long-term vision is to become a lightweight infrastructure component that AI engineers can integrate into existing RAG systems, AI agents, and LLM applications with minimal changes.

---

# The Problem

A typical Retrieval-Augmented Generation (RAG) pipeline looks like this:

```text
                 User Query
                      │
                      ▼
            Generate Embedding
                      │
                      ▼
          Search Vector Database
                      │
                      ▼
            Retrieve Knowledge
                      │
                      ▼
                Rerank Results
                      │
                      ▼
            Construct Prompt
                      │
                      ▼
                 Call the LLM
                      │
                      ▼
                  Final Answer
```

Even when multiple users ask nearly identical questions, this entire pipeline is often executed repeatedly.

As applications scale, this repeated computation becomes one of the largest contributors to latency and infrastructure cost.

---

# How Remem Works

Instead of assuming every request requires a full execution, Remem attempts to reuse previous work whenever it is safe to do so.

```text
                 Incoming Request
                        │
                        ▼
              Generate Embedding
                        │
                        ▼
               Semantic Similarity
                        │
                        ▼
            AI Work Reuse Decision
                        │
     ┌──────────────────┼──────────────────┐
     │                  │                  │
     ▼                  ▼                  ▼
Reuse Response   Reuse Retrieval     Execute Pipeline
     │                  │                  │
     │                  ▼                  ▼
     │           Skip Retrieval      Store New Execution
     │             & Reranking             │
     └──────────────────┴──────────────────┘
                        │
                        ▼
                 Return Result
```

Rather than acting as a traditional cache, Remem behaves like a decision engine that selects the highest level of reusable computation for each request.

---

# What Can Remem Reuse?

Depending on the request and available metadata, Remem may reuse:

* Entire LLM responses
* Retrieved knowledge chunks
* Reranked search results
* Prompt construction artifacts
* Tool execution results
* SQL query results
* Agent execution artifacts
* Future AI workflow outputs

Not every request can safely reuse every artifact.

For example:

* If the knowledge base has changed, Remem may skip response reuse but still reuse retrieval results.
* If the retrieval results have changed, Remem executes a fresh retrieval.
* If no previous work is reusable, the application executes normally.

Rather than guaranteeing that every request avoids an LLM call, Remem always chooses the **highest level of reusable work that preserves correctness**.

In some situations, this completely eliminates another LLM invocation.

In others, it may reuse only the retrieval stage while generating a fresh response.

This adaptive approach minimizes latency, reduces infrastructure cost, and maintains response quality without requiring changes to an application's existing retrieval pipeline.


# Design Philosophy

Remem follows a few simple principles.

## Correctness before optimization

Always build the correct solution before making it faster.

---

## Measure before optimizing

Every optimization should be supported by benchmarks.

---

## Keep public APIs simple

Internal architecture may evolve.

The public API should remain stable.

---

## Observe rather than replace

Remem should integrate with existing AI stacks instead of forcing engineers to adopt new databases or retrieval systems.

---

# Current Status

Current Version:

```
v0.6.0-alpha

```
Implemented so far:

- ✅ Decoupled Execution Contexts (ExecutionContext)

- ✅ Policy-driven compatibility checks (ReusePolicy)

- ✅ Metadata filtering before similarity math (MetadataMatcher)

- ✅ Modular folder structure for reuse layers

- ✅ Policy-compliant client interface

- ✅ Metrics Collector (MetricsCollector / MetricsSnapshot)

- ✅ Persistence Layer (durable JsonStorage + ephemeral InMemoryStorage)

- ✅ Pip-installable package (`pip install remem`)

Not yet implemented (Planned for future versions):

- Explanations API

- HTTP server

- Language SDKs (Java, Rust)

- Distributed mode

- Rust acceleration


---

# Current Architecture


The architecture cleanly decouples the public Client, policy engines, metadata matchers, and underlying storage/similarity layers, ensuring high modularity and scalability.

---

# Repository Structure

```
remem/
├── docs/
├── remem/
│   ├── models/
│   │   ├── execution_context.py
│   │   ├── execution_record.py
│   │   └── execution_result.py
│   ├── reuse/
│   │   ├── engine.py
│   │   ├── matcher.py
│   │   └── policy.py
│   ├── similarity/
│   └── storage/
├── examples/
├── benchmarks/
└── README.md
```

---

# Running the Examples

Clone the repository and install in editable mode (with dev extras):

```bash
git clone https://github.com/harshvardhansingh7/remem.git
cd remem
pip install -e ".[dev]"
```

Run the end-to-end RAG work-reuse example:

```bash
python examples/rag_reuse.py
```

Run the durable persistence example:

```bash
python examples/persistent_storage.py
```

---

# Running Tests

```bash
pytest
```

---


# Development Roadmap

## v0.1.0-alpha

- RetrievalEntry
- Similarity Engine
- In-memory Storage
- Unit Tests

---

## v0.2.0

Public Remem API

```
Remem()

↓

find_similar()

↓

Storage

↓

Similarity
```

---

## v0.3.0 

Intelligent Execution Reuse Engine with rich execution records (`get_or_compute`).

---

## v0.4.0 (Current)

Metadata-aware policy matching engine with explicit ExecutionContext and ReusePolicy.

---

## v0.5.0

Persistence Layer

Support durable storage instead of in-memory only.

---

## v0.6.0

Performance Optimization

- Faster similarity search
- Profiling
- Benchmarking
- Memory optimization

---

## v0.7.0

SDKs

- Python
- Java
- Rust

---

## v1.0.0

Production-ready AI Work Reuse Engine

---




# Long-Term Goals

Remem aims to support:

- Retrieval reuse
- Agent memory reuse
- Prompt context reuse
- SQL query reuse
- Tool execution reuse
- Knowledge versioning
- Distributed deployments
- High-performance storage engine
- Rust acceleration

---

# Why Not Just Use Redis?

Redis is an excellent key-value cache.

Remem solves a different problem.

Redis answers:

> "Have I seen this exact key before?"

Remem aims to answer:

> "Have I already performed similar expensive AI work before?"

Instead of exact key matching, Remem focuses on semantic similarity and work reuse.

---

# Learning Goals

Remem is also a personal engineering journey.

This project is being built from first principles to deeply understand:

- Distributed systems
- Storage engines
- Database internals
- Concurrency
- Memory management
- Networking
- Performance optimization
- AI infrastructure
- Open-source engineering

The goal is not simply to build another cache.

The goal is to build a useful infrastructure project while understanding every layer involved.

---

# Contributing

Contributions are welcome.

As the project is still in its early stages, architecture discussions and feedback are especially valuable.

Please read `CONTRIBUTING.md` before opening issues or pull requests.

---

# License

Licensed under the Apache License 2.0.

See the `LICENSE` file for details.

---

# Project Status

⚠️ **Early Alpha (v0.4.0)**

The project is under active development.

Breaking API changes are expected until the first stable release.

---

## Author

**Harshvardhan Singh**

Building Remem as an open-source AI infrastructure project to explore distributed systems, storage engines, and high-performance backend engineering.

If this project interests you, consider giving it a ⭐ and following its progress.
