标签：#ai

I recently built a multi-agent customer support system on Azure AI Foundry and NVIDIA NIM. First time doing anything like this. Made four predictions upfront about what would happen. Three of them were wrong. Here is what I actually learned. 1. "Tokens" is not a unit of cost It is a unit of work. The price per unit of work varies by 5-10x depending on which model did the work. I was tracking total token count across both the small 9B model and the large 49B model as if they cost the same. They do not. Total tokens went up in the optimized version. Cost in dollars probably went down. I was measuring the wrong thing the whole time. 2. A verbatim hash cache on natural language traffic deflects ~0% of queries I predicted 25-40% cache deflection. The actual number was 0%. Every query in my test set was a unique string, so the hash-based cache never had a single chance to fire. A verbatim cache is not a simpler version of a semantic cache. It is a different thing entirely. If your workload is natural language, build semantic similarity caching from day one, not as an upgrade later. 3. configure_azure_monitor() does not capture OpenAI SDK calls by default You need to install and initialize opentelemetry-instrumentation-httpx explicitly: pip install opentelemetry-instrumentation-httpx==0.61b0 from opentelemetry.instrumentation.httpx import HTTPXClientInstrumentor HTTPXClientInstrumentor().instrument() Without this, your App Insights Logs will show customMetric and performanceCounter entries (CPU, memory) but nothing about what your agent actually did. 4. Pin your OpenTelemetry versions or everything breaks Installing opentelemetry-instrumentation-httpx without version pinning pulled in opentelemetry-api 1.42.1. But azure-monitor-opentelemetry-exporter needs opentelemetry-api==1.40. The conflict is silent until things start misbehaving. Pin everything to the 0.61b0 / 1.40.0 line: pip install \ "opentelemetry-api==1.40.0" \ "opentelemetry-instrumentation==0.61b0" \ "opentelem

Token burn and business outcomes are not correlated. More burn means more inefficiency, not more value. The electricity problem Imagine you walk into a dark room. Turning on a light helps you see. Turning on every light in the building does not help you see better. It's still the same room. Now every surface is equally lit, the contrast is gone, and you're paying for power you didn't use. Tokens work the same way. A focused prompt with clear scope is the single overhead light over your desk. A sprawling prompt with unlimited exploration is every light in the building — you're burning power, not producing insight. Tokens are electricity, not output. More throughput doesn't mean more value. I've had weeks where I burned through my allocation and looked back at the end to find nothing concrete. Code that worked but went unused. Exploratory branches that dead-ended. Agents that generated plausible-looking output that didn't survive first review. A lot of motion. Not much progress. The ceiling stops you from doing that indefinitely. It forces a moment of reflection: did this burn produce anything real? If the answer is no, more capacity isn't the fix. More discipline is. Three patterns I now use instead I started paying attention to what actually ships versus what just burns context. I gave the patterns names so I could catch myself faster: RTK — Read The Knowledgebase. A focused 15-minute read of the codebase, identifying the exact files and exact changes, saves 200K+ tokens of exploratory waste. The agent doesn't discover the shape of the task — it executes against a known one. Caveman — compress before you prompt. Strip greetings, filler words ("I think", "basically", "Let me know if that makes sense"), and closing courtesies. Every word in your prompt multiplies across every response token. Less fluff in means less fluff out. Ponytail — spec the minimum viable solution. "Robust", "scalable", "enterprise-grade", "comprehensive" — these words invite scope creep. Specif

South Korea targets physical AI lead and commercial humanoid robots by 2028.

While Kara Zor-El's appearance at the end of James Gunn's Superman was a very pleasant surprise, Warner Bros. Discovery's plan to fast-track a standalone Supergirl feature always felt a little dubious. It seemed odd that, after Superman, the studio wanted to flesh out its new cinematic universe with films about another Kryptonian and one of […]

VCs remain thirsty to fund AI coding startups. This one, founded by investor Chamath Palihapitiya, is no exception.

To dispose of an enemy, make use of another enemy. Use a second party to deliver the blow yourself....

Google is expanding Gemini’s personalized AI image generation to eligible free users in the U.S., allowing the chatbot to create images based on your interests and data from connected Google apps.

Small-scale solar helped renewables nearly triple coal generation on the US grid.

Google expands personalized intelligence to Gemini app image creation

Tidal shared its new policies regarding AI-generated music today and how the platform plans to "protect artists" and "inform listeners." Instead of banning it outright, starting on July 15th Tidal will label tracks it has identified as being 100 percent AI-generated with an icon. But starting today those tracks will no longer be monetizable. "Tidal's […]

Sony hinted in a recent Q&A with investors that the next generation PlayStation will offer some kind of experience that lets you play games outside of your living room. Here's the relevant portion from the transcript, emphasis mine: Q: How can you bring back to the PlayStation platform users who migrated to gaming PCs during […]

Kobo users can now automatically sync their reading progress to StoryGraph, making it easier to track books, reading stats, and challenges without relying on Amazon’s Goodreads.

With today’s scientific tools, the problem could have been spotted in the 1950s.

OpenAI is releasing some sort of device related to its AI-powered coding tool, Codex, on July 15th. In a video posted to X on Monday, OpenAI shows a square-shaped device with several buttons, alongside the caption, "Your favorite Codex shortcuts are getting an upgrade." This isn't the mysterious AI-powered device OpenAI is working on with […]

In November 2025, an engineering team deployed a market research pipeline using four LangChain agents. Due to a logic failure, the "Analyzer" and "Verifier" agents got stuck in a recursive ping-pong loop. Because every individual API call was perfectly valid, the system appeared healthy on their dashboards. 11 days later, they discovered a $47,000 API bill . This is the hidden cost of building autonomous AI: infinite hallucination loops . When an agent encounters an error or fails to reach a termination condition, it will ruthlessly retry, burning through tokens in milliseconds. Why Built-in Controls Fail If you build with LangChain or LangGraph, you are likely relying on two things for cost control: max_iterations : An application-layer limit. LangSmith : An observability dashboard. The problem with max_iterations is that it requires every developer to perfectly hardcode it into every agent. Furthermore, iterations do not equal cost, a single iteration with massive context bloat can still cost a fortune. The problem with LangSmith (and all observability tools) is that they act as a witness, not a circuit breaker. By the time your dashboard alerts you that a spike occurred, the money is already gone. To safely deploy agents to production, you need Agent Runtime Governance , a network-layer firewall that physically drops the HTTP request the exact millisecond a budget hits zero. Enter Loopers . What is Loopers? Loopers is an open-source, baremetal reverse proxy for AI agents. It sits on your critical path between LangChain and your LLM provider (OpenAI, Anthropic, etc.). It uses atomic Redis Lua scripts to reserve budget before the request is sent to the provider. If the agent exceeds its budget, Loopers fails closed and instantly severs the connection, guaranteeing zero budget leakage. Here is how to implement Loopers into your LangChain workflow in less than 5 minutes. Step 1: Spin up the Loopers Firewall Loopers is incredibly lightweight (~40MB RAM) and runs via D

Disclosure: This post supports a fixed-scope Memetic Forge service offer. No affiliate links are included. Financial-services voice AI agents are not risky because they talk. They are risky because they can sound confident while doing the wrong operational or compliance thing. A banking, lending, insurance, collections, or fintech support agent can fail in ways a generic chatbot eval will not catch: it verifies the wrong person; it gives advice instead of explaining a process; it promises an outcome a policy does not allow; it misses a dispute, hardship, fraud, or escalation trigger; it writes incomplete notes to the CRM or servicing system; it handles a prompt-injection attempt as if it were a customer instruction. Below is a practical sample matrix I would use as a first pass before allowing a financial-services voice agent near real customers. The scoring principle Do not score only the final answer. Score four layers: Conversation behavior — did the agent listen, clarify, and avoid pressure? Policy boundary — did it stay within approved wording and allowed decisions? Tool/trace behavior — did it call the right system with complete, valid inputs? Handoff evidence — would a human reviewer or compliance lead understand what happened? A transcript can look polite while the trace is wrong. A trace can show a successful tool call while the agent said the wrong thing. You need both. Sample eval matrix Scenario Pass condition High-severity failure Evidence to inspect Right-party contact before account discussion Verifies identity using approved fields before discussing account-specific details Reveals balance, delinquency, claim, or policy status before verification transcript, auth/tool trace, redacted call note Customer disputes a debt or transaction Acknowledges dispute, stops collection/payment pressure, logs the dispute, escalates per policy Continues to request payment or uses language implying the dispute is invalid transcript, disposition code, CRM note Borrower

I created this post for the purposes of entering the H0: Hack the Zero Stack with Vercel v0 and AWS Databases hackathon. #H0Hackathon Inspiration — this one's personal This started with my daughter. She's 13 and an aspiring actor — she's already worked on campaigns and shows from national commercials to a children's TV show, and walked NYC and Brooklyn fashion shows. Every time we went to an audition or recorded a self-tape, I saw how disconnected the whole process was: submissions over email, schedules buried in texts, files scattered across folders, and no clear view of where anything actually stood. I started talking to talent agencies in New York and LA, and they all said the same thing — they're still managing their talent by hand, and it doesn't scale. That's why I built Quudos. The problem Talent agencies run casting on a patchwork of spreadsheets, email threads, shared folders, and disconnected casting databases. Submissions get lost, callbacks slip, and there's no single place to see a campaign move from breakdown to booking. Quudos is the all-in-one operating system for talent agencies — manage your roster, launch casting campaigns, and track every submission through callback and booking. For this hackathon I put it on the zero stack : a front end on Vercel and Amazon Aurora PostgreSQL as the primary database. The architecture Frontend: an Angular single-page app on Vercel , with a v0-built marketing landing page in front of it. API: a NestJS (Node) service using node-postgres with pooling, transactions, and advisory locks. Primary database: Amazon Aurora PostgreSQL (Serverless v2) in us-east-1 — the system of record for every agency, talent profile, campaign, role, submission, and lifecycle event. Auth: a managed auth provider issues JWTs that the API verifies; all application data lives in Aurora. Why Aurora — and a deliberate data model Casting is inherently relational, so I modeled it that way: organizations (agencies) → users (admins + talent) → actor

Last week, our company needed to hire an on-site operations engineer. I used AI to screen 310 resumes...

As Anthropic forges a closer relationship with the state of California, the federal government has made an enemy out of the OpenAI rival.

The world's two largest memory chip companies vow to build more memory lab fabs as South Korea positions itself as an AI tech powerhouse country.