Search posts, papers, and topics
Building AI agents is now easy, but as deployments grow into dozens or hundreds, companies lose visibility into ownership, usage, and cost—a problem called agent sprawl. Blocks Enterprise offers a private network that lets heterogeneous agents be discovered, permissioned, and audited across clouds without forcing a single framework.
The post explains how retrieval‑augmented generation (RAG) lets LLM‑based assistants answer questions from private corpora. It covers chunking documents into passages, embedding queries and chunks, similarity metrics, and the trade‑offs of different vector indexes (flat, IVF, HNSW). The focus is on practical design choices rather than new research.
Google's ADK for Kotlin 1.0 is a production-ready framework for building AI agents across Kotlin, Android, and JVM applications. It achieves feature parity with Python ADK, offering idiomatic Kotlin APIs for orchestration, tools, and memory, alongside new Android-specific on-device and hybrid AI capabilities.
Google’s Gemini Enterprise Agent Platform now offers a private‑preview Agent Anomaly Detection service that asynchronously analyzes agents’ OpenTelemetry traces and logs to flag risky behavior (e.g., tool misuse, privilege abuse) without adding latency, surfacing findings in Security Command Center and via an API for automated mitigation.
Nvidia frames AI security as an engineering discipline, outlining required controls across the AI agent stack (model, harness, runtime) and advocating for enforceable boundaries, traceable identities, and evidence‑based testing. It highlights OpenShell as a sandboxed runtime, the Open Secure AI Alliance, and several vendor tools for testing and red‑team exercises.
Nvidia’s Jensen Huang announced Salesforce’s Koa, a CRM‑reasoning LLM built by fine‑tuning Nvidia Nemotron 3 Super on a synthetic, three‑decade‑spanning dataset. Koa uses supervised fine‑tuning plus RL (NeMo RL, Gym, AutoModel), covers 14+ industries, and claims 3× fewer errors on Salesforce’s CRM‑Bench versus leading models. It’s already in internal Slack agents and slated for limited customer p…
Dropbox’s Riviera platform has been expanded from a preview service into a universal, composable content‑processing system that now handles hundreds of thousands of transformations per second across 300+ file types. The architecture separates orchestration from execution, adds a plugin model, and offers async public APIs so internal and external AI workloads can reuse cached transformations.
A vendor‑agnostic JSON Schema manifest for AI agent tools, with a TypeScript adapter layer that compiles the manifest into OpenAI function‑calling and Anthropic Claude formats, plus runtime validation, permission scoping, and execution constraints.
WSO2 Agent Manager GA adds model‑agnostic identity, role‑based access, policy enforcement (40+ controls), a Kubernetes‑native sandboxed runtime, and OpenTelemetry‑based tracing/evaluation for AI agents across cloud, on‑prem and hybrid environments.
Databricks built Marge, a Genie‑powered conversational analytics assistant on a governed Marketing Lakehouse. By starting with a single high‑value use case (email campaign performance), documenting data, encoding verified answers, teaching business terminology, and embedding the tool in existing ticket workflows, they achieved 85% adoption, 3× higher data usage in decisions, 50% QoQ usage growth,…
Microsoft open‑sourced TauGrid, a Helm‑installable platform that bundles Kueue, KubeRay and other components to schedule, monitor and checkpoint AI workloads on GPU‑enabled Kubernetes clusters. Engineers define jobs in a simple `tau.yaml`, and TauGrid handles queuing, health checks, and observability, simplifying AI workload management.
A hands‑on benchmark of five LLM code‑review models (Claude Opus 5, GPT 5‑6 Sol, Grok 4.6, Kimi 3, GLM 5.3, Qwen 3.8 Max, DeepSeek v4 Pro) shows that all models ace tiny 15‑line diffs but diverge sharply on 100‑300‑line, cross‑file changes. The author built a deterministic harness (single HTTP call, fixed system prompt, JSON output) and measured detection rate, false‑positives, latency, token usa…
Cloudflare’s Agent Development Lifecycle (ADLC) replaces the classic SDLC with an autonomous, event‑driven workflow system that lets AI agents handle code generation, testing, deployment, and maintenance. The platform builds on Cloudflare Workflows and the new @cloudflare/ci library to spin up containers, run headless browsers, and chain steps with caching and credential support. Observability is…
Yelp added an Inference Plugin to Nrtsearch that runs XGBoost and neural‑network models inside the search engine, eliminating a separate scoring service. The plugin extracts features from index documents, loads MLeap bundles from MLflow, and serves predictions on replica nodes with millisecond latency.
Databricks shows how to turn energy‑theft ML scores into a governed, end‑to‑end workflow using a Databricks App, Lakebase for live case state, Unity Catalog for data governance, and Genie One for natural‑language executive reporting. The pattern lets utilities act on alerts faster while staying compliant, and can be reused for other fraud‑type use cases.
Vercel AI Gateway now offers the GLM‑5.3‑FlashX model, a fast (~200 tps) multimodal coding LLM. The post includes a TypeScript streaming example, notes the model’s fit for coding agents and interactive tools, and lists AI Gateway features (unified API, usage tracking, retries/failover, custom reporting, key budgets, routing rules) with no platform fee.
Zapier’s blog post reviews six AI‑writing generators (Jasper, Anyword, Writer, Writesonic, Rytr, Sudowrite), explains how they work on top of LLMs, and outlines the author’s evaluation criteria (dedicated writing focus, control over output, ease of use, price, anti‑spam stance). The write‑up is a high‑level overview with no code, benchmarks, or deep technical analysis.
Nvidia’s blog spotlights five companies that are using Nvidia AI platforms to accelerate clean‑energy projects—from grid interconnection and nuclear plant operations to off‑grid AI data‑center power, advanced reactors, and fusion tokamaks. The article is a marketing summary and provides few technical details.
A marketing‑style roundup of eight AI‑powered sales assistant tools, describing high‑level features, pricing, and pros/cons without deep technical analysis or original evaluation.
A four‑stage model for adopting agentic AI in software delivery pipelines, warning that Stage 3 (bounded autonomous execution) is arriving faster than governance programs. The post explains the stages, required controls (scoped credentials, fail‑closed defaults, tamper‑evident logs), and uses the OpenClaw incident as a cautionary example.
Mozilla and Mila, backed by $5 M from Mozilla and $1 M from Hypertec plus Canadian government support, announce an open‑source AI foundation layer aimed at letting small organisations run private AI models. The post describes the high‑level vision (open standards + reference implementation) and a two‑year rollout plan, but provides no technical details, design rationale, or measurable results.
This is a 10‑question quiz that checks your grasp of installing Ollama, pulling models, choosing between chat and generate APIs, and managing multi‑turn conversations in Python. It reinforces the basics of running LLMs on your own hardware for privacy and offline use.
Qubika integrated Anthropic Claude (Chat, Code) and its Agentic Framework into Throne’s engineering workflow for bug fixes and feature work. Over the measured period they cut estimated effort by 52% and delivered code 2.1× faster while optimizing model selection for cost and latency.
AIM Research’s 2026 Databricks Services Partners report shows most partner work (50‑75% of projects) is still core lakehouse builds, data‑engineering modernization, and cloud migrations, with emerging focus on Unity Catalog governance, FinOps, and production‑grade agentic AI. Qubika ranks 5th in penetration (0.68) and 3rd in maturity (0.85), highlighted for its real‑time pipelines, reusable IP (Q…
