Vector database space is the Wild West, keep at it"},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Emerging architectures for LLM applications"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://a16z.com/2023/06/20/emerging-architectures-for-llm-applications/"}},"_tags":["comment","author_baner2022","story_36409489"],"author":"baner2022","children":[36412572],"comment_text":"Appreciate sharing, will try Milvus
Vector database space is the Wild West, keep at it","created_at":"2023-06-20T22:10:01Z","created_at_i":1687299001,"objectID":"36411392","parent_id":36410546,"story_id":36409489,"story_title":"Emerging architectures for LLM applications","story_url":"https://a16z.com/2023/06/20/emerging-architectures-for-llm-applications/","updated_at":"2024-09-20T14:22:27Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"shanghaikid"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"If you are not using GCP or you want to have an open-source alternative, Please check my project Milvus vector database (https://milvus.io).
We've published a bunch of demo cases powered by vector database on GitHub. https://github.com/milvus-io/bootcamp
We have built Milvus vector database upon ANN libraries like faiss, annoy, nsmlib, etc.
We are aiming to create a cloud-scalable vector database. So Milvus comes to the crossroad of vector search and cloud database. There are many interesting system design topics in the development of Milvus 2.0. We will continue to share our experiences and thoughts on this topic."},"story_title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["vector"],"value":"Find anything fast with Google's vector search technology"},"story_url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["vector"],"value":"https://cloud.google.com/blog/topics/developers-practitioners/find-anything-blazingly-fast-googles-vector-search-technology"}},"_tags":["comment","author_shanghaikid","story_29554986"],"author":"shanghaikid","comment_text":"If you are not using GCP or you want to have an open-source alternative, Please check my project Milvus vector database (https://milvus.io).
We've published a bunch of demo cases powered by vector database on GitHub. https://github.com/milvus-io/bootcamp
We have built Milvus vector database upon ANN libraries like faiss, annoy, nsmlib, etc.
We are aiming to create a cloud-scalable vector database. So Milvus comes to the crossroad of vector search and cloud database. There are many interesting system design topics in the development of Milvus 2.0. We will continue to share our experiences and thoughts on this topic.","created_at":"2021-12-15T03:50:20Z","created_at_i":1639540220,"objectID":"29561870","parent_id":29554986,"story_id":29554986,"story_title":"Find anything fast with Google's vector search technology","story_url":"https://cloud.google.com/blog/topics/developers-practitioners/find-anything-blazingly-fast-googles-vector-search-technology","updated_at":"2024-09-20T10:05:37Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"fzliu"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"Hey folks,
As much as we love GPT-4, it's expensive and can be slow at times. That's why we built GPTCache - a semantic cache for autoregressive LMs - atop the vector database Milvus and SQLite.
GPTCache provides several benefits:\n1) reduced expenses due to minimizing the number of requests and tokens sent to the LLM service\n2) enhanced performance by fetching cached query results directly\n3) improved scalability and availability by avoiding rate limits, and\n4) a flexible development environment that allows developers to verify their application's features without connecting to LLM APIs or network.
Come check it out! https://github.com/zilliztech/gptcache"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: GPTCache \u2013 Redis for LLMs"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/zilliztech/GPTCache"}},"_tags":["story","author_fzliu","story_35547426","show_hn"],"author":"fzliu","children":[35550755,35553408,35591558],"created_at":"2023-04-12T21:44:53Z","created_at_i":1681335893,"num_comments":5,"objectID":"35547426","points":7,"story_id":35547426,"story_text":"Hey folks,
As much as we love GPT-4, it's expensive and can be slow at times. That's why we built GPTCache - a semantic cache for autoregressive LMs - atop the vector database Milvus and SQLite.
GPTCache provides several benefits:\n1) reduced expenses due to minimizing the number of requests and tokens sent to the LLM service\n2) enhanced performance by fetching cached query results directly\n3) improved scalability and availability by avoiding rate limits, and\n4) a flexible development environment that allows developers to verify their application's features without connecting to LLM APIs or network.
Come check it out! https://github.com/zilliztech/gptcache","title":"Show HN: GPTCache \u2013 Redis for LLMs","updated_at":"2026-02-04T08:52:25Z","url":"https://github.com/zilliztech/GPTCache"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"pingsl"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"The background of the question:
We are developing an open-source cloud-scalable vector database: Milvus[1]. In our new cloud architecture, we need to employ a messaging framework to serve as the central log sequence of the whole system. Ultimately, we will support well-known messaging solutions like Apache Kafka, Apache Pulsar, etc. Referring to Confluent's article[2], Pulsar's message consumption model is push-based, while Kafka's is pull-based.
Our new cloud architecture is based on the actor model, which means all the worker nodes are working asynchronously, and the log sequence is the key to linking all the nodes. We think the push-based message consumption model and the actor model have more logical consistency. So we started with Apache Pulsar in our new implementation. We did have some concerns about Pulsar; for example,
- The project popularity
- The Go SDK is buggy
- The documentation is not as good as Kafka
We don't think these points are showstoppers. However, some of our users have different thoughts. They are asking when we would support Kafka. Some of them even mentioned they won't maintain Pulsar in their production environment.
So HN, please let me know your thoughts about this question. Should we give Kafka adoption a much higher priority since we just completed the new lease GA?
Reference:
[1] https://milvus.io
[2] https://www.confluent.io/kafka-vs-pulsar/"},"title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: Do you support Apache Kafka more than Apache Pulsar? Why?"}},"_tags":["story","author_pingsl","story_30100844","ask_hn"],"author":"pingsl","created_at":"2022-01-27T14:20:11Z","created_at_i":1643293211,"num_comments":0,"objectID":"30100844","points":4,"story_id":30100844,"story_text":"The background of the question:
We are developing an open-source cloud-scalable vector database: Milvus[1]. In our new cloud architecture, we need to employ a messaging framework to serve as the central log sequence of the whole system. Ultimately, we will support well-known messaging solutions like Apache Kafka, Apache Pulsar, etc. Referring to Confluent's article[2], Pulsar's message consumption model is push-based, while Kafka's is pull-based.
Our new cloud architecture is based on the actor model, which means all the worker nodes are working asynchronously, and the log sequence is the key to linking all the nodes. We think the push-based message consumption model and the actor model have more logical consistency. So we started with Apache Pulsar in our new implementation. We did have some concerns about Pulsar; for example,
- The project popularity
- The Go SDK is buggy
- The documentation is not as good as Kafka
We don't think these points are showstoppers. However, some of our users have different thoughts. They are asking when we would support Kafka. Some of them even mentioned they won't maintain Pulsar in their production environment.
So HN, please let me know your thoughts about this question. Should we give Kafka adoption a much higher priority since we just completed the new lease GA?
Reference:
[1] https://milvus.io
[2] https://www.confluent.io/kafka-vs-pulsar/","title":"Ask HN: Do you support Apache Kafka more than Apache Pulsar? Why?","updated_at":"2024-09-20T10:19:12Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"fzliu"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"Hey folks,
As much as we love GPT-4, it's expensive and can be slow at times. That's why we built GPTCache - a semantic cache for autoregressive LMs - atop the vector database Milvus and SQLite.
GPTCache provides several benefits:
1) reduced expenses due to minimizing the number of requests and tokens sent to the LLM service
2) enhanced performance by fetching cached query results directly
3) improved scalability and availability by avoiding rate limits, and
4) a flexible development environment that allows developers to verify their application's features without connecting to LLM APIs or network.
Come check it out! https://github.com/zilliztech/gptcache"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Reduce your GPT bill via semantic caching"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/zilliztech/GPTCache"}},"_tags":["story","author_fzliu","story_35630732","show_hn"],"author":"fzliu","created_at":"2023-04-19T17:16:41Z","created_at_i":1681924601,"num_comments":0,"objectID":"35630732","points":1,"story_id":35630732,"story_text":"Hey folks,
As much as we love GPT-4, it's expensive and can be slow at times. That's why we built GPTCache - a semantic cache for autoregressive LMs - atop the vector database Milvus and SQLite.
GPTCache provides several benefits:
1) reduced expenses due to minimizing the number of requests and tokens sent to the LLM service
2) enhanced performance by fetching cached query results directly
3) improved scalability and availability by avoiding rate limits, and
4) a flexible development environment that allows developers to verify their application's features without connecting to LLM APIs or network.
Come check it out! https://github.com/zilliztech/gptcache","title":"Show HN: Reduce your GPT bill via semantic caching","updated_at":"2024-09-20T13:48:29Z","url":"https://github.com/zilliztech/GPTCache"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ray927"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"Location: Hyderabad, India
Remote: Yes
Willing to relocate: Yes (for the right opportunity)
Technologies: Data Science, Machine Learning, Deep Learning, Python (for DS), R, AWS (EC2, Lambda, Sagemaker), GCP (Vertex AI), MLOps (MLFlow, W&B), SQL, NLP, Gen AI (Prompt Engineering, RAG), LLMs (PaLM 2, Llama 2, Gemma), ML Frameworks (Keras, Tensorflow, Flask, LlamaIndex, Langchain), Vector Databases (Milvus, ChromaDB), Technical Leadership, Team Management
R\u00e9sum\u00e9/CV: https://bit.ly/kiran-online-resume
Email: kvskiran92[at]gmail[dot]com
LinkedIn: https://bit.ly/kiran-linkedin
Experienced Data Scientist with 10+ years of total IT experience with 7+ years of hands-on experience in Data Science projects. Throughout my career, I worked on solving data science use cases in e-commerce(search, ads, ranking), consumer banking financial sector(risk analytics, fintech), and recruitment tech.
Passionate about solving complex problems with data and adding value to my team and company. Consistent with updating myself with emerging trends in the industry and willing to learn new technologies.
Actively seeking Lead or Staff Data Scientist full-time roles. Let's connect on LinkedIn. Thanks."},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: Who wants to be hired? (June 2024)"}},"_tags":["comment","author_ray927","story_40563280"],"author":"ray927","comment_text":"Location: Hyderabad, India
Remote: Yes
Willing to relocate: Yes (for the right opportunity)
Technologies: Data Science, Machine Learning, Deep Learning, Python (for DS), R, AWS (EC2, Lambda, Sagemaker), GCP (Vertex AI), MLOps (MLFlow, W&B), SQL, NLP, Gen AI (Prompt Engineering, RAG), LLMs (PaLM 2, Llama 2, Gemma), ML Frameworks (Keras, Tensorflow, Flask, LlamaIndex, Langchain), Vector Databases (Milvus, ChromaDB), Technical Leadership, Team Management
R\u00e9sum\u00e9/CV: https://bit.ly/kiran-online-resume
Email: kvskiran92[at]gmail[dot]com
LinkedIn: https://bit.ly/kiran-linkedin
Experienced Data Scientist with 10+ years of total IT experience with 7+ years of hands-on experience in Data Science projects. Throughout my career, I worked on solving data science use cases in e-commerce(search, ads, ranking), consumer banking financial sector(risk analytics, fintech), and recruitment tech.
Passionate about solving complex problems with data and adding value to my team and company. Consistent with updating myself with emerging trends in the industry and willing to learn new technologies.
Actively seeking Lead or Staff Data Scientist full-time roles. Let's connect on LinkedIn. Thanks.","created_at":"2024-06-05T15:27:32Z","created_at_i":1717601252,"objectID":"40586084","parent_id":40563280,"story_id":40563280,"story_title":"Ask HN: Who wants to be hired? (June 2024)","updated_at":"2024-09-20T17:15:58Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ray927"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"Location: Hyderabad, India
Remote: Yes
Willing to relocate: Yes (for the right opportunity)
Technologies: Data Science, Machine Learning, Deep Learning, Python (for DS), R, AWS (EC2, Lambda, Sagemaker), GCP (Vertex AI), MLOps (MLFlow, W&B), SQL, NLP, Gen AI (Prompt Engineering, RAG), LLMs (PaLM 2, Llama 2, Gemma), ML Frameworks (Keras, Tensorflow, Flask, LlamaIndex, Langchain), Vector Databases (Milvus, ChromaDB), Technical Leadership, Team Management
R\u00e9sum\u00e9/CV: https://bit.ly/kiran-online-resume
Email: kvskiran92[at]gmail[dot]com
LinkedIn: https://bit.ly/kiran-linkedin
Experienced Data Scientist with 10+ years of IT experience with 7+ years of hands-on experience in Data Science projects, currently leading a DS team. Kaggle Competitions Expert & Kaggle Kernels Expert.
Throughout my career, I worked on solving data science use cases in e-commerce(search, ads, ranking), consumer banking financial sector(risk analytics, fintech), and recruitment tech.
Passionate about solving complex problems with data and adding value to my team and company. Consistent with updating myself with emerging trends in the industry and willing to learn new technologies.
Actively seeking Lead or Staff Data Scientist full-time roles. Let's connect on LinkedIn. Thanks."},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: Who wants to be hired? (May 2024)"}},"_tags":["comment","author_ray927","story_40224210"],"author":"ray927","comment_text":"Location: Hyderabad, India
Remote: Yes
Willing to relocate: Yes (for the right opportunity)
Technologies: Data Science, Machine Learning, Deep Learning, Python (for DS), R, AWS (EC2, Lambda, Sagemaker), GCP (Vertex AI), MLOps (MLFlow, W&B), SQL, NLP, Gen AI (Prompt Engineering, RAG), LLMs (PaLM 2, Llama 2, Gemma), ML Frameworks (Keras, Tensorflow, Flask, LlamaIndex, Langchain), Vector Databases (Milvus, ChromaDB), Technical Leadership, Team Management
R\u00e9sum\u00e9/CV: https://bit.ly/kiran-online-resume
Email: kvskiran92[at]gmail[dot]com
LinkedIn: https://bit.ly/kiran-linkedin
Experienced Data Scientist with 10+ years of IT experience with 7+ years of hands-on experience in Data Science projects, currently leading a DS team. Kaggle Competitions Expert & Kaggle Kernels Expert.
Throughout my career, I worked on solving data science use cases in e-commerce(search, ads, ranking), consumer banking financial sector(risk analytics, fintech), and recruitment tech.
Passionate about solving complex problems with data and adding value to my team and company. Consistent with updating myself with emerging trends in the industry and willing to learn new technologies.
Actively seeking Lead or Staff Data Scientist full-time roles. Let's connect on LinkedIn. Thanks.","created_at":"2024-05-02T05:37:15Z","created_at_i":1714628235,"objectID":"40233024","parent_id":40224210,"story_id":40224210,"story_title":"Ask HN: Who wants to be hired? (May 2024)","updated_at":"2024-09-20T16:58:59Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ray927"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"Location: Hyderabad, India
Remote: Yes
Willing to relocate: Yes (for the right opportunity)
Technologies: Data Science, Machine Learning, Deep Learning, Python (for DS), R, AWS (EC2, Lambda, Sagemaker), GCP (Vertex AI), MLOps (MLFlow, W&B), SQL, NLP, Gen AI (Prompt Engineering, RAG), LLMs (PaLM 2, Llama 2, Gemma), ML Frameworks (Keras, Tensorflow, Flask, LlamaIndex, Langchain), Vector Databases (Milvus, ChromaDB), Technical Leadership, Team Management
R\u00e9sum\u00e9/CV: https://bit.ly/kiran-online-resume
Email: kvskiran92[at]gmail[dot]com
LinkedIn: https://bit.ly/kiran-linkedin
Experienced Data Scientist with 10+ years of IT experience with 7+ years of hands-on experience in Data Science projects, currently leading a small team. Kaggle Competitions Expert & Kaggle Kernels Expert.
Throughout my career, I worked on solving data science use cases in e-commerce(search, ads, ranking), consumer banking financial sector(risk analytics, fintech), and recruitment tech.
Passionate about solving complex problems with data and adding value to my team and company. Consistent with updating myself with emerging trends in the industry and willing to learn new technologies.
Actively seeking Lead or Staff/Senior Data Scientist full-time roles. Let's connect on LinkedIn. Thanks."},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Ask HN: Who wants to be hired? (April 2024)"}},"_tags":["comment","author_ray927","story_39894818"],"author":"ray927","comment_text":"Location: Hyderabad, India
Remote: Yes
Willing to relocate: Yes (for the right opportunity)
Technologies: Data Science, Machine Learning, Deep Learning, Python (for DS), R, AWS (EC2, Lambda, Sagemaker), GCP (Vertex AI), MLOps (MLFlow, W&B), SQL, NLP, Gen AI (Prompt Engineering, RAG), LLMs (PaLM 2, Llama 2, Gemma), ML Frameworks (Keras, Tensorflow, Flask, LlamaIndex, Langchain), Vector Databases (Milvus, ChromaDB), Technical Leadership, Team Management
R\u00e9sum\u00e9/CV: https://bit.ly/kiran-online-resume
Email: kvskiran92[at]gmail[dot]com
LinkedIn: https://bit.ly/kiran-linkedin
Experienced Data Scientist with 10+ years of IT experience with 7+ years of hands-on experience in Data Science projects, currently leading a small team. Kaggle Competitions Expert & Kaggle Kernels Expert.
Throughout my career, I worked on solving data science use cases in e-commerce(search, ads, ranking), consumer banking financial sector(risk analytics, fintech), and recruitment tech.
Passionate about solving complex problems with data and adding value to my team and company. Consistent with updating myself with emerging trends in the industry and willing to learn new technologies.
Actively seeking Lead or Staff/Senior Data Scientist full-time roles. Let's connect on LinkedIn. Thanks.","created_at":"2024-04-02T05:58:37Z","created_at_i":1712037517,"objectID":"39902800","parent_id":39894818,"story_id":39894818,"story_title":"Ask HN: Who wants to be hired? (April 2024)","updated_at":"2025-03-31T17:26:31Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"fzliu"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"Shameless self-plug for our embedded vector database milvus-lite (https://github.com/milvus-io/milvus-lite):
pip install milvus pip install milvusWe chose to build on PostgreSQL because of its readily available metadata search capabilities and proven scalability of relational databases. While PGVector has pioneered this approach, it\u2019s often perceived as slower than native vector databases like Milvus. Roo-VectorDB builds on the PGVector framework, incorporating our own optimizations in search strategies, memory management, and support for higher-dimensional vectors.
In preliminary lab testing using ANN-Benchmark, Roo-VectorDB demonstrated performance that was comparable to, or significantly better than, Milvus in terms of QPS (queries per second).
RooAGI will continue to develop AI-focused products, with Roo-VectorDB as a core storage component in our stack. We invite developers around the world to try out the current release and share feedback."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["vector"],"value":"Show HN: RooAGI's Roo-VectorDB: A New PostgreSQL Extension for Vector Search"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/RooAGI/Roo-VectorDB"}},"_tags":["story","author_rooagi","story_44572270","show_hn"],"author":"rooagi","children":[44609104],"created_at":"2025-07-15T15:37:00Z","created_at_i":1752593820,"num_comments":1,"objectID":"44572270","points":4,"story_id":44572270,"story_text":"RooAGI (https://rooagi.com) has released Roo-VectorDB, a PostgreSQL extension designed as a high-performance storage solution for high-dimensional vector data. Check it out on GitHub: https://github.com/RooAGI/Roo-VectorDB
We chose to build on PostgreSQL because of its readily available metadata search capabilities and proven scalability of relational databases. While PGVector has pioneered this approach, it\u2019s often perceived as slower than native vector databases like Milvus. Roo-VectorDB builds on the PGVector framework, incorporating our own optimizations in search strategies, memory management, and support for higher-dimensional vectors.
In preliminary lab testing using ANN-Benchmark, Roo-VectorDB demonstrated performance that was comparable to, or significantly better than, Milvus in terms of QPS (queries per second).
RooAGI will continue to develop AI-focused products, with Roo-VectorDB as a core storage component in our stack. We invite developers around the world to try out the current release and share feedback.","title":"Show HN: RooAGI's Roo-VectorDB: A New PostgreSQL Extension for Vector Search","updated_at":"2025-07-18T19:57:54Z","url":"https://github.com/RooAGI/Roo-VectorDB"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"bzGoRust"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["milvus","vector","database"],"value":"I would like to mention that vector databases like Milvus got lots of new features to support RAG, Agent development, features like BM25, hybrid search etc.."},"story_title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["database"],"value":"Databases in 2025: A Year in Review"},"story_url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["database"],"value":"https://www.cs.cmu.edu/~pavlo/blog/2026/01/2025-databases-retrospective.html"}},"_tags":["comment","author_bzGoRust","story_46496103"],"author":"bzGoRust","comment_text":"I would like to mention that vector databases like Milvus got lots of new features to support RAG, Agent development, features like BM25, hybrid search etc..","created_at":"2026-01-05T13:05:43Z","created_at_i":1767618343,"objectID":"46498341","parent_id":46496103,"story_id":46496103,"story_title":"Databases in 2025: A Year in Review","story_url":"https://www.cs.cmu.edu/~pavlo/blog/2026/01/2025-databases-retrospective.html","updated_at":"2026-03-05T23:19:56Z"}],"hitsPerPage":20,"nbHits":149,"nbPages":8,"page":0,"params":"query=Milvus+vector+database&advancedSyntax=true&analyticsTags=backend","processingTimeMS":28,"processingTimingsMS":{"_request":{"roundTrip":14},"afterFetch":{"format":{"highlighting":1,"total":1},"merge":{"mergeLoop":{"prepareNextHit":19,"total":19},"total":19},"total":20},"fetch":{"query":6,"total":7},"total":28},"query":"Milvus vector database","serverTimeMS":30}