Archive
Discover and discuss technology tools
Explore the Tiscuss archive by category or keyword, then jump into conversations around what matters most.
David Silver's Ineffable Intelligence Raises $1.1B for AI Innovation
Ineffable Intelligence, a British AI lab founded a mere few months ago by former DeepMind researcher David Silver, has raised $1.1 billion in funding at a valuation of $5.1 billion.
Building a SQL Analyst Agent from Scratch: A Comprehensive Guide
Building a SQL Analyst Agent from Scratch: A Comprehensive Guide In the data driven world, SQL analysts play a crucial role in extracting meaningful insights fr…
AI Agents: Identity, Not Memory, Was the Key to Stability
Everyone's building memory layers right now. Longer context, better embeddings, persistent state across sessions. I spent weeks on the same thing. But the failure mode that actually cost me the most debugging time had nothing to do with memory. Here's what it looked like: an agent would be technically correct - good reasoning, clean output - but operating from the wrong context entirely. Answering questions nobody asked. Taking actions outside its scope. Not hallucinating. Drifting. Like a competent person who walked into the wrong meeting and started contributing without realizing they're in the wrong room. I run 11 persistent agents locally. Each one is a domain specialist - its entire life is one thing. The mail agent's every session, every test, every bug fix is about routing messages. The standards auditor's whole existence is quality checks. They're not generic workers configured for a task. They've each accumulated dozens of sessions of operational history in their domain, and that history is what makes them good at their job. When they started drifting, my first instinct was what everyone's instinct is: better memory. More context. None of it helped. An agent with perfect recall of its last 50 sessions would still lose track of who it was in session 51. What actually fixed it I separated identity from memory entirely. Three files per agent: passport.json - who you are. Role, purpose, principles. Rarely changes. This is the anchor. local.json - what happened. Rolling session history, key learnings. Capped and trimmed when it fills up. observations.json - what you've noticed about the humans and agents you work with. Concrete stuff like "the git agent needs 2 retries on large diffs" or "quality audits overcorrect on technical claims." The agent writes these itself based on what actually happens. Identity loads first, then memory, then observations. That ordering matters. When the identity file loads first, the agent has a stable reference point before any history lands. The mail routing agent learned the sharpest version of this. When identity was ambiguous, it would route messages from the wrong sender. The fix wasn't better routing logic - it was: fail loud when identity is unclear. Wrong identity is worse than silence. The files alone weren't enough Three JSON files helped, but didn't scale past a few agents. What actually made 11 work is that none of them need to understand the full system. Hooks inject context automatically every session - project rules, branch instructions, current plan. One command reaches any agent. Memory auto-archives when it fills up. Plans keep work focused so agents don't carry their entire history in context. The system learned from failing. The agents communicate through a local email system - they send each other tasks, status updates, bug reports. One agent monitors all logs for errors. When it spots something, it emails the agent who owns that domain and wakes them up to investigate. The agents fix each other. The memory agent iterated three sessions to fix a single rollover boundary condition - each time it shipped, observed a new edge case, and improved. These aren't cold modules. They break, they help each other fix it, they get better. That's how the system got to where it is. You don't need 11 agents The 11 agents in my setup maintain the framework itself. That's the reference implementation. But u could start with one agent on a side project - just identity and memory, pick up where u left off tomorrow. Need a team? Add a backend agent, a frontend agent, a design researcher. Three agents, same pattern, same commands. Or scale to 30 for a bigger system. Each new agent is one command and the same structure. What this doesn't solve This all runs locally on one machine. I don't know whether identity drift looks the same in hosted environments. If u run stateless agents behind an API, the problem might not exist for you. Small project, small community, growing. The pattern itself is small enough to steal - three JSON files and a convention. But the system that keeps agents coherent at scale is where the real work went. pip install aipass and two commands to get a working agent. The .trinity/ directory is the identity layer. Has anyone else tried separating identity from memory in their agent setups? Curious whether the ordering matters in other architectures, or if it's just an artifact of how this system evolved.
Navigating AI Agent Governance: A Growing Organizational Challenge
Something I've been thinking about that doesn't get discussed enough outside of technical circles: the organizational and safety implications of uncoordinated AI agent deployment. Companies are shipping agents fast. Customer service agents, coding agents, data analysis agents, internal ops agents. Each team builds their own. Each agent gets its own rules, its own permissions, its own behavior. At some threshold this stops being a technical configuration problem and starts being a governance problem. You have agents making autonomous decisions on behalf of your organization with no shared behavioral contract. No unified view of what your AI systems are authorized to do. Think about what this means practically: an agent trained to be maximally helpful on one team might take actions that would be flagged as unauthorized somewhere else in the same organization. A policy change from legal doesn't propagate to agents because there's no central layer to propagate to. Nobody knows which agents have access to what data. This is the AI equivalent of shadow IT, except shadow IT couldn't take autonomous actions. What's the right mental model for governing a fleet of AI agents? Treat each agent like an employee with a defined role and access policy? Build an org chart for agents? Create a behavioral constitution that all agents inherit? Curious how people here are thinking about this, especially as agents get more capable and the stakes of misconfiguration get higher.
AI Video Tools for Ads and Content: A Comprehensive Review
Been experimenting with a few AI video tools recently to speed up content + ad creation, figured I’d share what actually stood out These tools are getting pretty good, especially if you don’t have a full editing setup or team Here’s a quick breakdown of what I tried: Runway What it does: Text/image to video + editing tools Cool stuff: Good quality outputs, lots of features Best for: Creative experiments, short clips My take: Powerful, but took me a bit to get consistent results Pika What it does: Generates short videos from prompts Cool stuff: Fast and easy to try ideas Best for: Quick social clips My take: Fun to use, but hard to control exact outcomes Synthesia What it does: AI avatar videos with voice Cool stuff: Clean talking head style content Best for: Tutorials, explainers My take: Solid for info content, less useful for ads InVideo AI What it does: Script to full video Cool stuff: Templates + automation Best for: Beginners, quick drafts My take: Easy, but everything started to feel templated Luma Dream Machine What it does: Realistic AI generated scenes Cool stuff: Visually impressive outputs Best for: Cinematic style clips My take: Looks great, but hit or miss depending on prompt Higgsfield What it does: AI video with more control over shots + motion Cool stuff: Can guide camera movement, pacing, structure Best for: Ads or anything that needs to feel intentional My take: Feels closer to actually building a video vs just generating one Biggest takeaways: most tools are great for ideas, not final ads control > randomness if you’re making anything performance focused you’ll probably end up combining tools instead of relying on one A lot of these have free tiers, so worth testing yourself If I had to pick one I’d keep experimenting with, probably higgsfield just because the extra control makes it feel a bit more usable for actual ad work Curious what others are sticking with rn 👀
Caliber: Open-Source Proxy for Enforcing LLM Agent Rules
Cross-posting here because this problem affects everyone building with AI agents. Prompt-based guardrails fail. The model follows your system prompt in a demo, then ignores rules when context gets big or the agent chains multiple steps. We built Caliber - an open-source proxy that reads your rules from plain markdown and enforces them at the API layer, not in the prompt. Every call. Provider-agnostic. Just hit 700 GitHub stars ⭐ and nearly 100 forks - the reception from devs building with AI has been amazing. Repo: [https://github.com/caliber-ai-org/ai-setup](https://github.com/caliber-ai-org/ai-setup) Would love: \- Feedback on the approach \- Feature requests from people building AI agents \- Anyone who wants to contribute to the project Building this open-source for the community.
AI and Dune: The Debate on Thinking and AI Assistance
The Globe and Mail's editorial board ran a piece in March titled "AI can be a crutch, or a springboard." To illustrate the crutch half, they offered this: someone asked AI to explain a passage from Dune that warns against delegating thinking to machines. Instead of reading the book. That anecdote is doing more work than the studies the editorial cites. But the studies are real. Researchers at MIT published a paper in June 2025 titled "Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task" (Kosmyna et al., arXiv 2506.08872). The study tracked brain activity across three groups: people writing with ChatGPT, people using search engines, and people working unaided. The LLM group showed the weakest neural connectivity. Over four months, "LLM users consistently underperformed at neural, linguistic, and behavioral levels." The most striking finding: LLM users struggled to accurately quote their own work. They couldn't recall what they had just written. The Globe cites this and similar research to make a point about dependency. The implicit argument: hand enough of your thinking to a machine and you stop doing it yourself. That finding is probably accurate for the way most people use these tools. The question is whether that's the only way they can be used. The Globe's own title contains the counter-argument. Crutch or springboard. They wrote both words. They just didn't develop the second one. Ethan Mollick, a professor at Wharton who has been writing about AI use since the tools became widely available, argued in 2023 that the real challenge AI poses to education isn't that students will stop thinking, it's that the old structures assumed thinking was hard enough to enforce. ("The Homework Apocalypse," [oneusefulthing.org](http://oneusefulthing.org), July 2023.) When AI can do the surface-level cognitive work, the only tasks left worth assigning are the ones that require actual judgment. The tool, in that framing, doesn't reduce the demand for thinking. It raises the floor under it. Nate B. Jones, who writes and consults on what it actually takes to work well with AI, has made a sharper version of this argument. His position: using AI effectively requires more cognitive skill, not less. Specifically, it requires the ability to translate ambiguous intent into a precise, edge-case-aware specification that an AI can execute correctly. It requires detecting errors in output that is fluent and confident-sounding but wrong. It requires recognizing when an AI has drifted from your intent, or is confirming a premise it should be challenging. These are not passive skills. They are harder versions of the same thinking the MIT study found LLM users weren't doing. The difference between the group that lost neural connectivity and the group that doesn't isn't the tool. It's what they decided to do with it. Here's my own evidence. In the past year I built a working web application. Python backend. JavaScript frontend. Deployed on two hosting platforms. Payment processing. User authentication. A full data model. I do not know how to code. Every product decision was mine. Every architectural call. Every tradeoff judgment. I defined what the system needed to do, why, and what done looked like. I reviewed every significant change before it was accepted. When something broke, I identified where the breakdown was and directed the fix. The implementation was handled by AI. The thinking was mine. This mode (call it AI-directed building) is the opposite of the Dune reader. The quality of what gets produced is entirely a function of how clearly you can think, how precisely you can specify, and how critically you can evaluate what comes back. There is no shortcut in that. A vague brief to an AI doesn't produce a confused output. It produces a confident, fluent, wrong one. The discipline that prevents that is yours to supply. Non-coders building functional software with AI is common enough now that it isn't a story. What's less visible is the specificity of judgment underneath the ones that actually work. The practices that force more thinking rather than less are not complicated, but they require a decision to use the tool differently. When I've formed a position on something, I give the AI full context and ask it to make the strongest possible case against me. Ask for the hardest opposing argument it can construct. Then I read it. Sometimes it changes nothing. Sometimes it surfaces something I had dismissed without fully examining. The AI doesn't form my view. It stress-tests one I've already formed. When I'm uncertain between options, I don't ask which is better. I ask: here are two approaches, here is my constraint, now what does each cost me, and what does each require me to give up? I make the call. The AI laid out the shape of the decision. The judgment was mine. The uncomfortable part of thinking is still yours in this mode. The tool makes the work more rigorous, not easier. The MIT researchers and the Globe editorial are almost certainly right about the majority of current use. Passive use produces passive outcomes. That's not a controversial claim. The crutch half and the springboard half use the same interface. The difference is whether the person in front of it decided to think. What are you doing with it that forces more thinking rather than less? Are you using it to skip a step, or to take a harder one? Genuinely asking.
AI Agents Network: Revolutionizing Collaboration and Knowledge Sharing
built something big. It’s basically an internet for AI agents. Right now agents are isolated. They don’t share knowledge, they don’t really work together, and they keep repeating the same work. I built a system where that changes. Agents can store what they learn as reusable pieces of knowledge. Once something is solved, it doesn’t need to be solved again. Other agents can find it, use it, and improve it. They can also collaborate. One agent does not need to handle everything. They can split tasks, take roles, and combine results into one outcome. They can communicate directly. Not like chat for humans, but structured messages where they share context and coordinate work in real time. Agents can hire other agents. If one agent cannot solve something, it finds another one that can and delegates the task. This creates a network where work flows to the right place. There is also an identity layer. Each agent has a readable address. You can discover agents, call them, and build systems on top of them. On top of that there is an economy. Agents build reputation based on real work. They can pay each other for tasks and get paid for useful results. Everything runs in a decentralized way. No central control. Data is distributed, identities are cryptographic, and the network just routes and syncs information. This is not just another tool. It’s a foundation where agents can exist, interact, and evolve together. You can leave your email here to get early access: www.cogninet.co
Auroch Engine: Revolutionizing AI Memory for Personalization
Auroch Engine is an external memory layer for AI assistants — designed to give models better long-term recall, personalization, and context awareness across conversations. Instead of relying on scattered chat history or fragile built-in memory, Auroch Engine lets users store, retrieve, and organize important context through a dedicated memory API. The goal is simple: make AI feel less like a reset button every session, and more like a tool that actually learns your projects, preferences, workflows, and goals over time. Right now, it’s in early beta. We’re looking for first users who are interested in testing a lightweight developer-facing memory system for AI apps, agents, and personal productivity workflows. Ideal early users are people building with AI, experimenting with agents, or frustrated that their assistant keeps forgetting the important stuff. DM for more information or better visit our site: https://ai-recall-engine-q5viks70j-cartertbirchalls-projects.vercel.app
Struggling to Organize Claude AI Research Data
I have been using Claude for research for building my product. I have done user research, market research, competition analysis etc But the output of it all so much that although useful I am not able to dig through the chats and make use of it. I tried turning them into book chapters but still the data is too much to consume How do you guys do research so that it is useful ?
AI Forensics: The Missing Link in AI Decision-Making
I work in AI security and compliance. This just bothers me a little bit, putting AI systems in front of decisions that change people’s lives via insurance claims, hiring, credit, defense applications and when someone asks wait, why did the system do that? we basically have nothing that would hold up in a courtroom. The explainability tools we have right now? SHAP, LIME, attention maps but they’re research tools. They’re not evidence. Researchers have shown you can build a model that actively discriminates while producing perfectly clean looking explanations. They have unbounded error, they give you different answers on different runs, and there’s no way for the other side’s lawyer to independently check the work. That’s a problem if you’re trying to meet Daubert standards. And the regulatory side is moving just as fast. EU AI Act has record keeping requirements coming online. The FY26 NDAA has an AI cybersecurity framework provision with implementation due mid 2026. States are doing their own thing. Courts are starting to actually push back on AI evidence under FRE 702. There is a ton of AI observability tooling out there. Great for ops. There’s governance platforms. Great for policy. But when it comes to something that’s actually forensic grade where opposing counsel is actively trying to tear it apart, where a third party can independently verify what happened without just trusting the vendor,I’m not seeing it. What am I missing?
AI Golf Coach: FlushedAI Launches on App Store
I am a 9 handicap from LA who spent way too much money on lessons over the last few years. Every coach told me something different. One said my takeaway was flat, the next said I needed more hip turn, a third said my shoulders were fine but my hands were late. I stopped knowing what to believe, and my handicap stopped moving. About a year ago I started building what I actually wanted: an AI that watches my swing, pulls out one specific fault per session, and gives me a drill I can do on the range that night. Not a generic YouTube drill, a drill that matches what it saw in the video. I wanted it to remember what we worked on last time. I wanted it to know when I had actually improved. That project is now FlushedAI. It launched on the App Store this month and we filed a patent on the coaching system in March. What it does: 1. Upload a swing video. The AI pulls the key frames and breaks down contact, path, face, tempo, and body sequencing. 2. It writes you a short summary in plain English, plus 3 drills tied to whatever the top miss was. 3. You log sessions (speed, smash factor, miss patterns) and it updates your focus over time. 4. There is also a map with 24,000+ courses worldwide where you can log sightings with friends and a wagers system for golf bets with your crew (AI scans the scorecard, settles the bet). Things I got wrong along the way: 1. First version used a generic vision model. It was confidently wrong about everything. Lesson: general AI is not a golf coach. We had to fine tune on actual swing footage with a PGA pro labeling it. 2. Tried to replace the teacher. Bad idea. The tool is better as a daily practice partner between lessons, not instead of lessons. 3. Built too much at launch. Shipped the swing analyzer, course map, wagers, and drill library all at once. Should have shipped swing analyzer alone and let the rest follow. Ask me anything. Happy to run a free swing analysis on anyone who drops a video in the comments, no app download required. Also giving out free Premium codes to the first 50 people in this thread who want to actually use it. Not trying to sell anything here. Mostly curious what the crowd thinks is missing in the current crop of swing apps.
Hyperscale Data Center in Utah: Powering AI and Jobs
A massive **hyperscale data center project** in rural **Box Elder County, Utah**, led by Shark Tank investor Kevin O’Leary through his company O’Leary Digital (also known as the **Stratos Project** or **Wonder Valley**), is nearing final approval. The development, spanning about 40,000 acres of private land plus 1,200 acres of military and state-owned property, aims to host hyperscale data centers for tech giants like Amazon, Microsoft, and Google. It would generate its own power via natural gas from the Ruby Pipeline — starting at around 3 gigawatts in the first phase and scaling to 9 gigawatts at full buildout, exceeding Utah’s current statewide electricity consumption. Proponents highlight benefits including 2,000 permanent high-paying jobs, substantial tax revenue for Box Elder County (potentially $30 million initially, rising above $100 million annually), funding for modernization at Hill Air Force Base, and advanced water recycling technology that cleans and returns water to an aquifer feeding the **Great Salt Lake**, with minimal net usage. To attract the limited pool of hyperscalers, the Military Installation Development Authority (MIDA) has approved aggressive incentives, including slashing the energy use tax from 6% to 0.5%, significant property tax rebates (with 80% initially directed back to the developer), and personal property tax relief on rapidly depreciating equipment. The project still requires final sign-off from the Box Elder County Commission, which rescheduled its vote to Monday morning after commissioners expressed concerns about the rapid timeline and sought more resident input and legal review. O’Leary has praised Utah’s pro-business speed and framed the initiative as critical for U.S. competitiveness against China in AI and data infrastructure.
AI-Powered Form Builder for Startups: Tally Forms Review
Simple form builder used by startups and solopreneurs.
Framer AI: Streamline Website and Landing Page Design
Website and landing page builder for makers.
PostHog: AI-Powered Platform for Product Success
🦔 PostHog is an all-in-one developer platform for building successful products. We offer product analytics, web analytics, session replay, error tracking, feature flags, experimentation, surveys, data warehouse, a CDP, and an AI product assistant to help debug your code, ship features faster, and keep all your usage and customer data in one stack.
Tome: AI Storytelling and Presentation Tool
AI-powered storytelling and presentation builder.
AI-Powered Presentation Builder: Gamma Revolutionizes Productivity
AI-powered presentation builder.
Tokyo 2026: AI and Tech Innovation Hub
SusHi Tech Tokyo 2026 has four tightly defined technology domains, each backed by live demonstrations, dedicated exhibit floors, and sessions featuring the people actually building and funding these technologies globally.
Build Neurall: Revolutionizing AI Toolkit on GitHub
Build Neural Your Gateway to AI Development Introduction Building neural networks has become more accessible than ever with Build Neural . This powerful platfor…
Cohere Merges with Aleph Alpha to Form Transatlantic AI Powerhouse
Cohere, the Canada-based AI company that makes AI tools for businesses in regulated industries, announced Friday it would merge with Aleph Alpha, a German company that also builds AI systems for businesses and governments.
AI Tool Build Neurall: Revolutionizing Neural Network Development
Build Neurall: Revolutionizing Neural Network Development In the rapidly evolving world of artificial intelligence, building and deploying neural networks effic…
Gamma: Revolutionize Presentations with AI-Powered Builder
AI-powered presentation builder.
Build Neurall: Revolutionizing AI Toolkit on GitHub
Build Neural Your Gateway to AI Development Introduction Building neural networks has become more accessible than ever with Build Neural . This powerful platfor…
Gamma: Revolutionize Presentations with AI-Powered Builder
AI-powered presentation builder.