AI & MLAdvanced
AI Agent Architecture Patterns for .NET: Orchestration, Memory and A2A
Design AI agents in .NET: single vs multi-agent, sequential, concurrent, handoff, group chat and Magentic orchestration, memory, approvals, durability and A2A.
Large Language Models
78 articles about Large Language Models: in-depth .NET and AI guides, senior interview questions and AI news on DotNet AI Hub.
Technologies 14
AI & MLAdvanced
Evaluating AI Applications in .NET: Quality, Safety and CI
Evaluate LLM apps in .NET with Microsoft.Extensions.AI.Evaluation: quality and safety evaluators, golden datasets, LLM-as-judge, caching, reports and CI gates.
AI & MLIntermediate
Azure OpenAI and Microsoft Foundry for .NET Developers
Use Azure OpenAI in Microsoft Foundry (formerly Azure AI Foundry) from .NET: deployments, keyless auth, guardrails, quotas, PTUs, Agent Service and cost.
AI & MLIntermediate
Function Calling and Tool Use with LLMs in C#
Master function calling in C# with Microsoft.Extensions.AI: AIFunctionFactory, FunctionInvokingChatClient, parallel calls, approvals, security and tests.
AI & MLAdvanced
Observability and Cost Control for LLM Apps in .NET
Learn observability and cost control for .NET LLM apps, covering OpenTelemetry GenAI conventions, token tracking, caching, model routing and 429 retries.
AI & MLAdvanced
Local AI in .NET: ONNX Runtime, Ollama and Foundry Local
Run LLMs locally in .NET with ONNX Runtime GenAI, Ollama and Foundry Local: small language models, GPUs and NPUs, quantization and IChatClient.
AI & MLIntermediate
Microsoft.Extensions.AI: Unified AI Abstractions for .NET
Master Microsoft.Extensions.AI: IChatClient, IEmbeddingGenerator, streaming, tool calling, middleware pipelines, DI registration, custom clients and testing.
AI & MLIntermediate
Multimodal AI in .NET: Vision, Audio and Speech
Build multimodal .NET apps that understand images, documents, speech and audio, and generate images and voice, with Microsoft.Extensions.AI and Azure AI.
AI & MLIntermediate
Using the Official OpenAI .NET Library for C#
A practical guide to the OpenAI .NET SDK: ChatClient, streaming, tool calls, structured outputs, the Responses API, embeddings, images, audio and retries.
AI & MLBeginner
Prompt Engineering for .NET Developers
A practical guide to prompt engineering for .NET developers, covering prompt anatomy, few-shot examples, reasoning models, templates and injection risks.
AI & MLAdvanced
Retrieval-Augmented Generation (RAG) in .NET: End-to-End Guide
Build production RAG in C#, covering ingestion, chunking, embeddings, hybrid retrieval, reranking, grounded prompts, citations and evaluation in .NET.
AI & MLIntermediate
Responsible AI and LLM Security for .NET Applications
Secure LLM apps in .NET: OWASP Top 10 for LLMs 2026, prompt injection defenses, Prompt Shields, PII redaction, output checks, least privilege and the EU AI Act.
AI & MLIntermediate
Semantic Kernel in Practice: Plugins, Functions and Orchestration
Semantic Kernel in practice for C#: kernel, plugins, prompt templates, automatic function calling, filters, vector search, agents, and its 2026 role.
AI & MLIntermediate
Structured Outputs: Reliable JSON from LLMs in C#
Get reliable JSON from LLMs in C#: JSON mode vs structured outputs, typed Microsoft.Extensions.AI responses, JSON Schema from types, validation and retries.
Interview Questions 5
AI EngineeringArchitect
AI Agents and MCP Interview Questions
Architect-level interview questions on AI agents and MCP: the agent loop, tool design, orchestration, memory, human-in-the-loop and failure modes in .NET.
AI EngineeringArchitect
Productionizing AI Applications Interview Questions
Architect-level interview questions on productionizing AI: evaluation, OpenTelemetry GenAI observability, cost, latency, prompt injection and compliance.
AI EngineeringSenior
LLM Integration Interview Questions for .NET Developers
Senior .NET interview questions on Microsoft.Extensions.AI, streaming, function calling, structured output, provider abstraction, retries and testing LLM code.
AI EngineeringArchitect
RAG and Vector Search Interview Questions
Architect-level interview questions on RAG: chunking, embedding choice, hybrid search, reranking, groundedness, stale data, permissions and cost control.
System DesignArchitect
System Design Interview: Enterprise AI Assistant (RAG Platform)
Architect-level system design questions on an enterprise RAG platform in .NET: ingestion, retrieval, security trimming, caching and cost control.
AI News 59
Claude Agents Find a New Enzyme System With CRISPR-Like Repeats
Anthropic said on September 23, 2026 that Claude agents found ART, a phage enzyme system with CRISPR-like repeats, as it unveiled a new life sciences lab.
Anthropic Launches Claude Opus 5.5 at 40% Lower Cost Than Opus 5
Anthropic released Claude Opus 5.5 on September 22, 2026 at $4/$20 per million tokens, with Fable-class safeguards and four breaking API changes to plan for.
Judge Rules Pentagon's Supply Chain Risk Label on Anthropic Unlawful
A federal judge ruled the Pentagon's supply chain risk designation of Anthropic unlawful. Here is how the dispute unfolded and what it means for AI buyers.
Alibaba Releases Qwen3.8, Its First Qwen-Max-Class Open-Weight Model
Alibaba's Qwen team released Qwen3.8 in August 2026, its first Qwen-Max-class open-weight model: a 2.4T MoE with 95B active parameters, followed by a 27B model.
Moonshot AI Releases Kimi K3 Weights, a 2.8-Trillion-Parameter Open Model
Moonshot AI released Kimi K3's weights in July 2026, a 2.8T-parameter multimodal MoE model with a 1M-token context and new terms for large commercial users.
Anthropic Says It Has Never Sought a Ban on Open-Weights Models
Dario Amodei said on July 27, 2026 that Anthropic has never sought a ban on open-weights models and backs chip controls and safety tests for all capable models.
Anthropic Launches Claude Sonnet 5, Its Most Agentic Sonnet Model
Anthropic released Claude Sonnet 5 on June 30, 2026, an agentic model close to Opus 4.8 at $2/$10 per million tokens and the new default on Free and Pro plans.
SpaceX to Acquire Cursor for $60 Billion in a Record All-Stock Deal
SpaceX agreed to buy AI coding startup Cursor for $60 billion in stock and closed the deal in August. What it means for developers who rely on Cursor.
SpaceX Debuts on Nasdaq in a Record IPO, Taking xAI Public With It
SpaceX, which absorbed xAI in February, priced its record IPO at $135 and closed its first day near $161. What the listing means for the AI industry.
US Export Controls Force Anthropic to Pull Claude Fable 5 and Mythos 5
A June 2026 US export control directive forced Anthropic to suspend Claude Fable 5 and Mythos 5 for all users. What happened and why it matters.
Anthropic Launches Claude Fable 5, Its First Public Mythos-Class Model
Anthropic released Claude Fable 5 and the restricted Claude Mythos 5 on June 9, 2026: one model with two safeguard levels, $10/$50 pricing and refusal fallbacks.
OpenAI Confidentially Files for an IPO, One Week After Anthropic
OpenAI confidentially filed for an IPO on June 8, 2026, a week after Anthropic, at a last valuation of $852B. What a listing could mean for developers.
Microsoft Unveils Seven In-House MAI Models at Build 2026
At Build 2026 on June 2, Microsoft launched seven in-house MAI models, led by the MAI-Thinking-1 reasoning model and the MAI-Code-1-Flash coding model.
Anthropic Confidentially Submits Draft S-1, Opening the Door to an IPO
Anthropic confidentially submitted a draft S-1 to the SEC on June 1, 2026, days after a $965B valuation. What an Anthropic IPO could mean for developers.
Anthropic Raises $65B Series H at a $965B Valuation, Passing OpenAI
Anthropic raised $65 billion at a $965 billion valuation, topping OpenAI, as run-rate revenue passed $47 billion. What it means for Claude developers.
Google Launches Gemini 3.5 Flash at I/O 2026 for Agents and Coding
Google opened its Gemini 3.5 series at I/O on May 19, 2026, with Gemini 3.5 Flash, which it says beats Gemini 3.1 Pro on agent benchmarks, at $1.50/$9 pricing.
Google Spots First Zero-Day Exploit It Believes Was Built With AI
Google's threat intelligence group found criminals preparing mass exploitation with a zero-day exploit it believes AI helped build. Key lessons.
Anthropic Signs Deal for All of SpaceX's Colossus 1 Compute Capacity
On May 6, 2026, Anthropic agreed to use all compute at SpaceX's Colossus 1 data center, over 300 MW and 220,000 NVIDIA GPUs, and raised Claude usage limits.
DeepSeek V4 Released: Open-Weight 1.6T MoE With a 1M-Token Context Built for Agents
DeepSeek released V4-Pro and V4-Flash in April 2026, MIT-licensed open-weight MoE models with a 1M-token context and far cheaper long-context inference.
OpenAI Releases GPT-5.5, Its Strongest Agentic Coding Model Yet
OpenAI released GPT-5.5 on April 23, 2026, with top agentic coding scores and GPT-5.4-level latency, then brought it to the API at $5/$30 per million tokens.
Google Unveils TPU 8t and TPU 8i, Its Eighth-Generation AI Chips
Google announced TPU 8t and TPU 8i at Cloud Next on April 22, 2026, splitting its eighth-generation TPU into separate chips for training and inference.
Anthropic Commits Over $100 Billion to AWS for Up to 5 GW of Compute
On April 20, 2026, Anthropic committed over $100 billion over ten years to AWS for up to 5 GW of compute, and Amazon agreed to invest up to $25 billion more.
Anthropic Launches Claude Managed Agents, a Hosted Runtime for AI Agents
Anthropic put Claude Managed Agents into public beta on April 8, 2026, a hosted agent runtime with sandboxes, long-running sessions and credential vaults.
Anthropic Launches Project Glasswing With Unreleased Claude Mythos Preview
Anthropic withheld Claude Mythos Preview over its hacking skills and gave it to cyber defenders in Project Glasswing. What it means for patching.
Google Releases Gemma 4 Open Models Under Apache 2.0, From Phones to a Single GPU
Google DeepMind released Gemma 4 in April 2026, multimodal open models from E2B to 31B, and moved Gemma to the Apache 2.0 license for the first time.
OpenAI Closes a $122 Billion Funding Round at an $852 Billion Valuation
OpenAI raised $122 billion at an $852 billion valuation from Amazon, Nvidia, SoftBank and retail investors. What the record round means for developers.
Mistral Small 4: Apache 2.0 Model Unifies Reasoning, Coding and Vision
Mistral AI released Mistral Small 4 in March 2026, an Apache 2.0 open-weight 119B MoE model with 6.5B active parameters that unifies chat, reasoning and coding.
OpenAI Launches GPT-5.4 with Native Computer Use and 1M-Token Context
OpenAI released GPT-5.4 on March 5, 2026, with native computer use, tool search and a 1M-token API context, priced at $2.50 and $15 per million tokens.
Google Releases Gemini 3.1 Pro in Preview with Big Reasoning Gains
Google released Gemini 3.1 Pro in preview on February 19, 2026, reporting 77.1% on ARC-AGI-2, adding a medium thinking level and keeping $2/$12 pricing.
Anthropic Releases Claude Opus 4.6 with a 1M-Token Context Window
Anthropic's Claude Opus 4.6 (February 5, 2026) added a 1M-token context in beta, adaptive thinking and agent teams, while keeping $5/$25 pricing.
OpenAI Launches a Standalone Codex App for Managing Coding Agents
OpenAI launched a standalone Codex app for Mac on February 2, 2026, built to help developers run and manage several AI coding agents at once. What changed.
SpaceX Merges With xAI in a $1.25 Trillion All-Stock Deal
SpaceX acquired Elon Musk's xAI in an all-stock merger valuing the combined company at $1.25 trillion. What the deal means for AI developers and compute.
OpenClaw One-Click RCE Flaw Puts Self-Hosted AI Agent Security in Focus
A one-click RCE flaw in the viral OpenClaw agent, plus malicious skills, showed the risks of self-hosted AI agents. What happened and how to run agents safely.
Microsoft Unveils Maia 200, an Inference-First AI Accelerator for Azure
Microsoft unveiled Maia 200 on January 26, 2026, a 3nm Azure inference chip it says offers 30% better performance per dollar than its existing systems.
OpenAI Signs $10 Billion-Plus Deal for 750 MW of Cerebras Compute
OpenAI signed a multiyear deal on January 14, 2026 for 750 MW of Cerebras computing power, reportedly worth over $10 billion, to speed up AI inference.
Apple Picks Google Gemini to Power Its Next Foundation Models and Siri
Apple and Google agreed that Gemini will underpin the next Apple Foundation Models and a more personalized Siri. What the partnership means for developers.
Google Launches Gemini 3 Pro, Topping LMArena and Debuting Antigravity
Google released Gemini 3 on November 18, 2025, starting with Gemini 3 Pro in preview, which topped LMArena at 1501 Elo and launched with Antigravity.
Whisper Leak: Microsoft Finds Encrypted LLM Streams Can Reveal Chat Topics
Microsoft's Whisper Leak research shows encrypted, streamed LLM responses can reveal conversation topics through packet sizes and timing. Key facts and fixes.
Anthropic Releases Claude Sonnet 4.5, Its Top Model for Coding Agents
Anthropic's Claude Sonnet 4.5, released September 29, 2025, led SWE-bench Verified and OSWorld at Sonnet pricing and arrived with the Claude Agent SDK.
Anthropic Endorses California SB 53, the Frontier AI Transparency Bill
Anthropic endorsed California's SB 53 on September 8, 2025. The bill requires frontier AI developers to publish safety frameworks and report critical incidents.
Anthropic Bars Chinese-Controlled Companies From Claude, Wherever They Operate
Anthropic tightened its regional rules on September 4, 2025, barring entities over 50% owned by firms in unsupported regions such as China from using Claude.
Anthropic Raises $13B Series F at a $183B Valuation, Led by ICONIQ
Anthropic closed a $13 billion Series F at a $183 billion valuation as run-rate revenue topped $5 billion. What the round signaled for Claude developers.
Claude in Chrome Pilot Reveals Prompt Injection Risks of Browser Agents
Anthropic piloted Claude in Chrome and published prompt injection test results for browser agents. What the numbers show and how developers should respond.
OpenAI Launches GPT-5, a Unified Model That Decides When to Think
OpenAI released GPT-5 on August 7, 2025, merging fast answers and deep reasoning in one system, with new API controls, lower prices and fewer hallucinations.
OpenAI Releases gpt-oss-120b and gpt-oss-20b, Its First Open-Weight Models Since GPT-2
OpenAI released gpt-oss-120b and gpt-oss-20b in August 2025, its first open-weight language models since GPT-2, under Apache 2.0 for reasoning and tool use.
Anthropic Says It Will Sign the EU General-Purpose AI Code of Practice
Anthropic said on July 21, 2025 that it will sign the EU General-Purpose AI Code of Practice, days before AI Act duties for model providers began to apply.
Moonshot AI Open-Sources Kimi K2, a 1-Trillion-Parameter Agentic Model
Moonshot AI open-sourced Kimi K2 in July 2025, a 1-trillion-parameter MoE model with 32B active parameters built for tool use, under a modified MIT license.
xAI Launches Grok 4 and Grok 4 Heavy, Claiming Top Reasoning Scores
xAI released Grok 4 and the multi-agent Grok 4 Heavy on July 9, 2025, with a 256K-token API priced at $3 and $15 per million tokens and strong benchmark claims.
Gemini 2.5 Pro and Flash Reach General Availability on Google Cloud
Google made Gemini 2.5 Pro and 2.5 Flash generally available on June 17, 2025, previewed 2.5 Flash-Lite and gave developers stable thinking models.
Claude Code Is Generally Available, With IDE Extensions and an SDK
Anthropic made Claude Code generally available on May 22, 2025, adding VS Code and JetBrains extensions, GitHub Actions support and the Claude Code SDK.
Anthropic Launches Claude Opus 4 and Claude Sonnet 4 for Coding Agents
Anthropic's Claude Opus 4 and Sonnet 4 (May 22, 2025) led coding benchmarks, added tool use during extended thinking and made Claude Code generally available.
Microsoft.Extensions.AI and Vector Data Extensions Reach General Availability
Microsoft declared Microsoft.Extensions.AI and the Vector Data extensions generally available in May 2025, giving .NET a stable, provider-neutral AI layer.
Alibaba Releases Qwen3: Apache 2.0 Open Models with Hybrid Thinking Modes
Alibaba's Qwen team released Qwen3 in April 2025: eight Apache 2.0 open-weight models, from 0.6B to a 235B MoE, that switch between thinking and fast modes.
OpenAI Releases o3 and o4-mini, Reasoning Models That Use Tools
OpenAI's o3 and o4-mini, released April 16, 2025, reason with tools and images, posted strong coding and math results and reached developers through the API.
OpenAI Launches GPT-4.1 in the API with a 1M-Token Context Window
OpenAI released GPT-4.1, mini and nano in April 2025: API-first models with a 1M-token context, stronger coding and more literal instruction following.
Meta Releases Llama 4 Scout and Maverick, Its First Open-Weight MoE Models
Meta released Llama 4 Scout and Maverick in April 2025, its first natively multimodal mixture-of-experts open-weight models, with a 10M-token context claim.
Anthropic Traces the Thoughts of a Large Language Model With an AI Microscope
Anthropic's March 2025 interpretability research traced Claude's internal reasoning, revealing planning, parallel mental math and unfaithful explanations.
OpenAI Adopts the Model Context Protocol, Starting With the Agents SDK
On March 26, 2025, OpenAI said it would support Anthropic's Model Context Protocol across its products, starting with the Agents SDK. What it means for devs.
DeepSeek-R1 Release: MIT-Licensed Open Reasoning Model Rivals OpenAI o1
DeepSeek-R1 arrived in January 2025 as an MIT-licensed open-weight reasoning model that DeepSeek says matches OpenAI o1, and it shook AI markets within a week.