Chris Bartholomew (@cdbartholomew) header
Chris Bartholomew (@cdbartholomew) profile picture

Building Hindsight @Vectorizeio | Creator of @langstream_ai | Prev founder @kesquetech (acquired @DataStax) | Author @OReillyMedia

Joined August 2011
Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)Photo by Chris Bartholomew (@cdbartholomew)
Another big milestone for Hindsight! A tiny fraction of repos on GitHub every reach this level. Thanks to the @Vectorizeio team, our many contributors, and the community. A special thanks to @nicoloboschi , who has lead the way on the project and the community.
1
9
555
Chris Bartholomew retweeted
Hindsight just hit 60k stars on Github!
2
1
15
518
Chris Bartholomew retweeted
best memory for Hermes, people are saying it, not me
15
10
3
109
15,945
Chris Bartholomew retweeted
2
5
192
Chris Bartholomew retweeted
I just upgraded my local AI memory tonight. Hindsight 1.2.0 + GPT OSS 120B + hs-decide-1-flash handling recall, consolidation, and reflect. Hs-decide-1-flash is now assisting in Recall, Reflect, and Consolidation, so the big LLM gets 31% fewer LLM calls. Who else is running a persistent agent memory layer locally? What are you using? I have been VERY happy with #Hindsight. I hope these changes were worth it.
4
3
7
195
Chris Bartholomew retweeted
this is the beginning of something new in the Memory space not only Hindsight can use decision models to get better in all the axis (cost, speed, quality) but we also train our own models now and we give them away FOR FREE mark my words: other memory systems will follow this train and copy us (as usual)
Introducing Hindsight Models Hindsight memory is now 2.5x cheaper and 2x faster, with better answers. The LLM writes, a small decision model judges. Open weights on Hugging Face, or hosted on the new Hindsight Inference API.
2
2
15
593
I believe that in the future all software will be powered by custom-trained models. Because these models are purpose-trained, they can be smaller and faster than general-purpose models. Today Hindsight takes a step in that direction with our own set of models trained specifically to improve how Hindsight works. I am also happy to report that these models also do well on industry benchmarks. The models are open source, but we are launching inference API if you don't want the hassle of self hosting them.
Introducing Hindsight Models Hindsight memory is now 2.5x cheaper and 2x faster, with better answers. The LLM writes, a small decision model judges. Open weights on Hugging Face, or hosted on the new Hindsight Inference API.
3
1
11
513
Dear Cold Email Bots, Asking me a "Quick Question" is a dead give away. You gotta come up with something better. You are AI, I know you can do it if you try.
2
1
120
Chris Bartholomew retweeted
Replying to @DaveShapi
Have you tried hindsight? On a setep like Hermes? If you haven't, your missing out on an incredible memory system.
3
5
81
Nailed it
i asked Claude to build the best memory system then this happened
5
136
Chris Bartholomew retweeted
started using Hermes 2 days ago today i'm installing tools i didn't even know existed just to make my agent stronger @Teknium your tool turned me into a nerd and there's no going back 💀
16
7
1
84
6,927
Hindsight: big name in memory for Hermes
Memory is the one Hermes setup everyone argues about. So here are 24 ways to give Hermes a memory, all in the official plugin catalog. Cloud, fully offline, plain notes or shared across every AI you use. Pick yours. The big names ▸ Hindsight github.com/vectorize-io/hind… ▸ Honcho github.com/plastic-labs/honc… ▸ mem0 github.com/mem0ai/mem0 ▸ OpenViking github.com/volcengine/OpenVi… ▸ Supermemory github.com/supermemoryai/her… ▸ Cognee github.com/topoteretes/cogne… Local and private ▸ Mnemosyne github.com/mnemosyne-oss/mne… ▸ Holographic github.com/itpartypattaya/he… ▸ Compartment github.com/MaxFreedomPollard… ▸ SuperLocalMemory github.com/qualixar/superloc… ▸ ReMe github.com/agentscope-ai/ReM… ▸ HyAtlas github.com/tuancookiez-hub/H… Notes, logs and wikis ▸ hwiki github.com/ottosulin/hwiki ▸ Memory Wiki (official) github.com/NousResearch/herm… ▸ GBrain github.com/kerrz2020/hermes-… ▸ OptMem github.com/rarf/optmem-herme… ▸ LCM-X github.com/electricsheephq/l… ▸ hermes-dreaming github.com/itpartypattaya/he… Cloud, shared and coding ▸ Memanto github.com/moorcheh-ai/meman… ▸ BetterDB github.com/BetterDB-inc/moni… ▸ Memoria github.com/matrixorigin/memo… ▸ Membase github.com/unibaseio/membase ▸ deja-memory github.com/vshulcz/deja-vu ▸ Sugar github.com/roboticforce/suga… Save this for the next time someone asks which memory to use.
1
12
719
Chris Bartholomew retweeted
The #1 long-term memory for Hermes? A folder of notes. 138 Hermes users answered: ▸ Obsidian / Markdown vault: 27 ▸ Hindsight: 25 ▸ Honcho: 22 ▸ Built their own: 18 25+ tools named. Local beat cloud 28 to 2. Memory is still anyone's game. Every system ranked below.
What are you using for long-term memory in your Hermes setup right now? Built-in memory, Honcho, Mem0, Hindsight, something else? What made you choose it?
26
28
1
462
29,470
Chris Bartholomew retweeted
Replying to @nvidia
@nvidia Nemotron 3.5 Embed is the new top embedding model on the Hindsight leaderboard. 93.8% recall@10 on LoCoMo, 65.9% on BEAM 100K, and the highest MRR of anything we have run end to end. Thanks to the @nvidia Nemotron team for the early access:
2
9
427
Chris Bartholomew retweeted
toktok is the first library to have a @typesafeai jev tokenizer github.com/vectorize-io/tokt…
3
5
288
Chris Bartholomew retweeted
Upgraded my OpenClaw to 2026.10.4, and one of the biggest surprises was Hindsight 1.0.0. Turns out @Vectorizeio Hindsight, which I've been using as my agent's long-term memory, now integrates with OpenClaw's native memory provider API. I've been running Hindsight for a while now. I even built my own MCP server for recall and reflect because I wanted my agent to actually make use of those memories. Now OpenClaw can route memory.search and memory.get directly to Hindsight. Which means I can finally start getting rid of some of my hand-rolled memory integration code. lol What I really like is that Hindsight is now treated as a first-class memory provider, instead of relying entirely on custom plugin hooks and MCP tools. As someone who has spent way too much time wiring AI tools together, that's a pretty nice upgrade 😂
4
3
1
10
428
Chris Bartholomew retweeted
Guys I'm so Hindsight pilled by @Vectorizeio I have a memory bank for @RuneScape 3 If you played RS3 you'll know it's a huge game, and I'm too busy to read the wiki. So instead I ask questions to my agent, specially about questing and where to get a certain item. And then hindsight saves it all. Hindsight also tracks what I do, as I move on from quest to quest, it kinda creates these decision records, and it even tells me "you were here the last time", or "you got that item that other day ago it's probably in your bank". If you're not using something like this in 2026 bro, this is like having a second brain. I love it. This is super cheeky but I even have an @NousResearch Hermes bot for it! I call it Zaros which is the god of fate and control in Runescape. I want so hard to make an RS3 plugin for this, but I don't have time! Runescape 3 + Hindsight + Hermes is amazing!
1
5
1
10
292
Chris Bartholomew retweeted
Replying to @HermesWatcher
Hindsight is AMAZING
2
2
14
2,194