Still wasting hours digging through folders and PDFs?
Your AI Avatar instantly learns your documents and website to build a flawless knowledge base.
Fast, accurate answers that stay on-brand and protect your message.
Hire your smartest team member via the link in bio! #AIAssistant #KnowledgeBase #SmartBusiness
2
Agritourism is helping rural communities diversify agricultural economies by allowing farms to host public-facing uses like education, recreation, and on-site retail. It supports farmers while shaping how rural character is managed.
Explore the Agritourism Knowledgebase Collection for policy guidance, planning tools, and examples from communities across the country.
planning.org/knowledgebase/a…
83
Most chatbots guess when they don't know.
BestChatBot answers only from your knowledge base, cites the docs and URLs it used, and says when the answer isn't in your sources.
Try it free: bestchatbot.io
#AIChatbot #KnowledgeBase
1
4
If your name is your business: I'm looking for you. This is for experts, creators and professionals that are looking to optimize their existing knowledgebase products.
Serious inquiries only. You must already have a course, ebook, newsletter or other digital product(s) available and ready to be seen as the obvious choice in your space. If you're still figuring out what you want to be known for, this isn't for you.
1
2
54
- Build voice AI agents on Bolna in under an hour
- Supports multiple agent types (LLM, knowledgebase, graph, etc.)
- @cartesia TTS makes it feel smoooth.
- Low latency = smooth, real-time experience
1
1
17
Fucking 100 iq retards talking about shit, going beyond their knowledgebase..
28
Not kidding, I saw a project where .md files were been linted, diff was being studied and everytime an agentic debugging session happened, it actually committed to the knowledgebase. Then a scheduled cron would sync them all.
Knowledge Engineering is a must if you really wanna work on enterprise codebases with agents. The entire workflow lives in the Knowledge Base including the troubleshooting steps and bugs.
3
Chaotic dev day, but LuciAIOS still pulled new users real engagement.
New build includes: – Cybersecurity add‑on – Luxury automation add‑on – Eco‑System integration – Knowledgebase engine – OS‑level control – Smart‑home layer – Dynamic voice upgrade
Momentum continues.
Onward
7
Not an "honorary" degree, a real doctorate. The word "doctor" comes from Latin, meaning "teacher". It means she is the utmost authority in her exact field of research, by producing original results that advance our knowledgebase. MD's are physicians.
1
34
My domain is automotive. Broadly. Too lazy to go into detail.
My approach is inefficient, but it's what works for me, at least so far.
If I understand correctly, they only "learn" within a context window, so I typically use Claude as a thinking partner. This develops thorough context. I generate a working document throughout which details constraints, goals, and decisions made including why those decisions were made.
I rarely accomplish this in one session, so the working document serves to provide the summarized content built from the last session. Most project require 3 or 4 sessions.
When it comes to coding, Claude then has a clear picture, but I still have to make it present the architecture and get approval from me before generating code.
Even with the context built from prior sessions, the first shot is almost always overly complex.
So I guess to sum up, I provide thorough context within a session, and then micromanage. It's much better at writing code than I am, but much worse at defining the architecture.
If you like, I will share a current working document. It's not top secret, but I don't want to post it for the entirety of X to see, so you'll have to send me a message.
Oh...one more thing. I have a knowledgebase which is just a bunch of .md files detailing various subjects. I often refer Claude to it so I don't have to type the same damn thing over and over.
1
39
That's where I need your knowledgebase, training and binary logic circuits.
Is Zuck taking us to the stars or just reorganising connectivity and efficiency in a ruthlessly monetisable way?
Infer my intent from input language and state your baseline on answer.
1
4
Jun 13
ALT A customer called today. "I've been billed $299 for 8 months. My contract says $199." I checked their account. In our billing system: $299. In the CRM: $199. In the signed contract PDF: $199. In the finance spreadsheet: $299. In the customer's email confirmation: $199. I asked finance. "We updated pricing in January. The system auto-upgraded existing customers." I asked sales. "We promised existing customers keep their rate." I asked the customer. "I have an email from your sales rep saying my rate is locked." We've been overcharging them by $100/month for 8 months. That's $800. Plus the time to fix it. Plus the trust we lost. Nobody documented which customers were grandfathered. So we grandfathered nobody.
Autonomous vehicles are expected to reshape how cities and regions plan for transportation, with levels of automation already being tested on public streets.
Explore the Autonomous Vehicles Knowledgebase Collection for background, policy guidance, and examples of how AV adoption may impact the built environment and transportation systems.
planning.org/knowledgebase/a…
1
356
How many times have you said:
"I know we solved this before."
Neural Nexus helps teams find old solutions, decisions and technical knowledge instantly.
🔗 neural-nexus-gules.vercel.ap…
#AItools #StartupFounder #KnowledgeBase #Tech
2
Jun 13
Knowledgebase of how to physically build everything, and a wearable robot suit to teach the techniques back to the next generations
11
Jun 13
The second part is not correctly judged, IMHO. Your assumption is:
generateContent(two-line prompt, LLM)
It does happen, and yes, slop content produced in this way is frustrating.
But there's a more typical scenario:
generateContent(prompt, LLM, my private knowledgebase)
Giving you the prompt alone would not be sufficient. Giving you my private knowledgebase is not feasible.
There can indeed be inequity in care given v/s care expected. But the private KB does contribute to the "care given". I think, for now, we have to consider slop on a case-by-case basis.
1
1
1
412
great idea here - (1) The problem: conversations are where work actually happens, and they evaporate the moment they end. Relevnt closes the loop. Clip on a mic pendant: a prep brief before the meeting, live nudges during, a speaker-labeled transcript, follow-up draft, and a playful courtroom verdict after. Everything lands in a team knowledgebase you chat with.
Jun 12
Our team won Best Use Case of Exa AI at SuperAI NEXT 2026. 36 hours at Marina Bay Sands, one wearable second brain for real conversations. The build:
(1) The problem: conversations are where work actually happens, and they evaporate the moment they end. Relevnt closes the loop. Clip on a mic pendant: a prep brief before the meeting, live nudges during, a speaker-labeled transcript, follow-up draft, and a playful courtroom verdict after. Everything lands in a team knowledgebase you chat with.
(2) The Exa part that won the prize: a Scout agent researches the person you are about to meet using Exa people search and content retrieval, disambiguates similar names, then a Briefer agent compresses the dossier into a one page brief plus a goal checklist. After the call, the pipeline runs Exa again to enrich every person extracted from the transcript. Walk in prepared, walk out with structured memory.
(3) The pipeline: a DJI Mic Mini inside a 3D printed pendant, Amazon Transcribe streaming over a WebSocket, a Live Copilot watching the rolling transcript, then a batch pass with speaker labels, summary, topics, and follow-up. RAG runs on Titan embeddings Nova generation with a grounding gate that drops any claim it cannot cite.
(4) Four people, four lanes: one ran the entire AWS backend, CDK, Lambdas, DynamoDB, end to end. Two ran the deck and the demo. Design and frontend made up the last lane, with Claude Code carrying most of the architecture. First hackathon shipped as a team instead of solo, and trust turned out to be the real velocity unlock.
(5) Underrated discovery: the Exa team is cracked. Some of the brightest, highest-energy people at the venue, helping teams debug like it was their own submission. And the API matched the people: search and content retrieval just worked for 36 straight hours.
(6) Most teams lose the hackathon in the last three minutes, not the first 33 hours. Judges experience a demo, not your commit history. Build less, rehearse more.
(7) A rule written down for my future self, for every team build: never share AI output you have not read and verified yourself. Unread AI output is spam with extra steps.
Repo and architecture in the reply.
49
Just in case you missed it: Gleap rolls out YouTube Video Analysis 👉 aho.is/bd91ad59 #customersupport #knowledgebase #AI
2
11
Jun 12
Locked context between iterations. If the agents previous state survives the update and the eval gates still pass, the rest is just a redeploy. Most breakage is the agent losing what it knew. You’ve gotta secure your living knowledgebase
3