Resources

High-performance code editor built in Rust with native AI integration, Agent Panel for agentic editing, and the open-source Zeta edit prediction model.

Agentic IDE with Cascade agent for multi-step code edits and full codebase indexing, acquired by Cognition (maker of Devin) in July 2025.

Experiment tracking, model versioning, and collaboration platform handling 10,000+ concurrent experiments with expanding LLM and agent evaluation capabilities.

Google Cloud agent platform combining managed Agent Engine deployment with the open-source ADK framework for code-first multi-agent development across Python, TypeScript, Java, and Go.

The leading TypeScript toolkit for building AI-powered applications and agents, with a provider-agnostic unified API for streaming, tool use, and agent abstractions.

Cloud FinOps platform with first-class AI cost management, native Anthropic and OpenAI cost ingestion, an MCP Server for natural-language cost queries, and an autonomous FinOps Agent.

MicroVM-based code execution environment integrated with Together AI model inference for end-to-end agent workflows on a single platform.

High-performance inference for 200+ open-source LLMs with sub-100ms latency, automated optimization, and horizontal scaling at lower cost than proprietary solutions.

Daily byte-sized summaries of the most important AI and machine learning news. Covers research papers, product launches, and industry developments in a 5-minute read for 500k+ subscribers.

Gergely Orosz's widely-read newsletter covering software engineering practices, tech industry trends, and engineering management. One of the most trusted sources for understanding how top engineering organizations operate and adopt new technologies including AI.

Daily AI news digest that cuts through the hype to deliver actionable insights. Known for its concise format and focus on practical AI developments.

DeepLearning.AI's weekly newsletter curated by Andrew Ng. Covers the most important AI news, research papers, and applications with thoughtful analysis and commentary.

GitHub's open-source toolkit for spec-driven AI development. Provides templates, a CLI, and prompts that center work around a specification-first workflow — from specs to technical plans to small, testable tasks that AI agents implement.

Data-native AI agent platform orchestrating across structured and unstructured data sources with Cortex Analyst for SQL queries, Cortex Search for documents, and AISQL for declarative inference pipelines.

Minimalist agent library from Hugging Face where agents write Python code as actions instead of JSON tool calls, with sandboxed execution via E2B, Docker, or WebAssembly.

In-depth technical writing on LLMs, prompt engineering, and AI tooling from the co-creator of Django. Known for practical explorations of new AI capabilities and building tools like LLM and Datasette.

Salesforce AI agent platform with Agent Script language, 60+ MCP tools, native CRM data access, and voice capabilities for building autonomous agents across sales, service, and marketing.

High-performance vector similarity search engine written in Rust with ACID-compliant transactions, metadata filtering, and distributed deployment.

Type-safe, model-agnostic agent framework from the Pydantic team with FastAPI-style developer experience, validated inputs/outputs, and OpenTelemetry instrumentation.

Prompt management platform with version control, automated evaluations, observability, A/B testing, and a no-code interface for non-technical team members.

Making artificial intelligence practical, productive, and accessible to everyone. Hosted by Daniel Whitenack and Chris Benson, the show covers AI topics from a practitioner perspective.

Enterprise AI gateway routing to 200+ LLMs with integrated guardrails, real-time cost observability, semantic caching, and automatic provider optimization.

Fully managed vector database with sub-50ms query times, automatic scaling, and serverless architecture for RAG applications.

Open-source LLM evaluation and observability platform from Comet with automated prompt optimization for agent workflows.

A lightweight, open-source spec-driven framework that works across 30+ AI coding agents. Generates proposals, technical designs, and implementation tasks from change descriptions, creating persistent context that survives chat sessions.

Unified API gateway providing access to 290+ AI models through a single OpenAI-compatible endpoint with dynamic cost and speed routing variants.

Open-source AI coding agent for terminal, IDE, and desktop. Supports 75+ LLM providers including local models, multi-session parallel agents, and LSP integration. Over 100k GitHub stars.

Lightweight multi-agent framework with four core primitives — agents, handoffs, guardrails, and tracing — available in Python and TypeScript.

Practical exploration of AI's impact on work, education, and daily life by Ethan Mollick, Wharton professor and author of the NYT bestseller "Co-Intelligence". Grounded in research, focused on what you can actually do.

Programmable guardrails toolkit for LLM systems supporting content moderation, PII detection, jailbreak prevention, and topic filtering via the Colang scripting language.

Serverless cloud platform for spinning up sandboxed containers with GPU access, autoscaling from zero to thousands of concurrent executions.

Open-source platform for the full ML lifecycle — experiment tracking, reproducibility, deployment, and model registry — with expanding LLM tracking and evaluation features.

Self-hosted microVM sandboxing platform for AI agent code execution with hardware-level isolation, sub-200ms boot times, and MCP integration.

TypeScript-native agent framework from the Gatsby team with model routing to 40+ providers, built-in evaluations, and observability.

In-depth technical YouTube channel and podcast featuring long-form interviews with leading AI researchers. Known for rigorous, no-hype discussions on machine learning theory, AI safety, and the philosophical implications of artificial intelligence.

Data framework for connecting LLMs to your data with sophisticated indexing, retrieval, and agent workflow capabilities including document OCR and multi-modal content.

Open-source Python SDK and AI Gateway providing a unified OpenAI-compatible interface to 100+ LLM APIs with budget controls, cost tracking, and load balancing.

Modern project management tool built for software teams. Known for its speed, keyboard-first design, and opinionated workflow that reduces overhead. Increasingly used alongside AI coding agents for issue tracking and sprint management.

Comprehensive technical deep-dives on machine learning concepts by Lilian Weng, Head of Applied Research at OpenAI. Posts are extensively researched and serve as definitive references.

The podcast by and for AI Engineers. Hosted by swyx and Alessio, featuring deep technical conversations with leading AI practitioners, researchers, and founders building the future of AI applications.

Full-lifecycle agent observability and evaluation platform from LangChain with multi-turn conversation scoring and automated behavior analysis.

Graph-based framework for building stateful, multi-actor agent applications with nodes, edges, and conditional routing for controllable orchestration.

Open-source LLM engineering platform for tracing, evaluation, prompt management, and granular cost analytics. Acquired by ClickHouse in 2025.

The most widely-adopted framework for building LLM-powered applications. Provides composable components for chains, agents, retrieval, and memory, plus LangGraph for complex multi-agent orchestration and LangSmith for observability.

Real-time AI security API for prompt injection detection, data loss prevention, and jailbreak prevention with sub-50ms latency.

Kubernetes-native controller providing a declarative CRD API for managing isolated, stateful pods purpose-built for AI agent code execution.

Weekly deep analysis of cutting-edge AI research papers with policy and safety implications, written by Jack Clark, co-founder of Anthropic. Read by 96k+ subscribers who want to understand where AI is heading.

Enterprise AI agent platform combining watsonx.ai for model access and agent development with watsonx Orchestrate for multi-agent workflows, featuring 150+ pre-built agents and hybrid cloud deployment.

Open-source LLM observability proxy for monitoring, evaluating, and optimizing cost and latency across providers with response caching and intelligent routing.

Python framework for LLM output validation with a Hub of pre-built validators, automatic retry and correction, and deployable as a standalone REST API.

Ultra-fast AI inference using custom Language Processing Units (LPUs) delivering up to 18x faster inference than traditional GPUs for latency-critical applications.

Microsoft's open-source data pipeline that extracts structured knowledge graphs from unstructured text using LLMs. Enables more sophisticated retrieval-augmented generation by reasoning over graph structures rather than flat document chunks.

Code-first, modular framework for building and deploying AI agents in Python, TypeScript, Java, and Go, optimized for Gemini but model-agnostic.

Cloud development environment platform (rebranded as Ona in 2025) now focused on orchestrating teams of AI software engineering agents in secure, ephemeral workspaces.

AI pair programmer integrated into VS Code, JetBrains, and Neovim with Agent Mode for autonomous multi-file editing and self-healing error correction.

Steve Yegge's multi-agent workspace manager that coordinates 20-30 parallel Claude Code agents through persistent git-backed state, role-based specialization (Mayor, Polecats, Crew), and the Beads issue tracking system.

AI reliability platform using purpose-built Luna small language models for sub-200ms hallucination detection and real-time production guardrails.

Fast model inference API with proprietary FireAttention engine supporting text, image, and audio with HIPAA and SOC2 compliance.

Fast-paced, highly entertaining tech content that makes complex topics accessible. Known for \"X in 100 seconds\" format covering AI, web development, and emerging technologies.

AI-focused publication by Dan Shipper serving 100k builders. Produces in-depth essays on AI implementation, the 'AI & I' podcast, and original AI productivity tools like Spiral and Cora.

Open-source Firecracker microVM sandboxes purpose-built for running AI-generated code in isolated environments with 150ms cold starts.

Pytest-native LLM evaluation framework with 50+ research-backed metrics, agent trace evaluation, and red-teaming capabilities.

Open-source infrastructure for running AI-generated code with sub-90ms sandbox creation, stateful execution, and programmatic file system and Git APIs.

Extension of Datadog monitoring for end-to-end AI agent tracing with automatic cost calculation, infrastructure cost correlation, and dedicated agent monitoring console.

Unified data and AI platform with Mosaic AI Agent Framework for building agents on enterprise data, Agent Bricks for auto-optimization via synthetic data, and MLflow 3 for GenAI lifecycle management.

AI-powered code editor built on VS Code. Features intelligent code completion, chat-based coding assistance, and codebase-aware suggestions that understand your entire project context.

Python framework for orchestrating role-based AI agents that collaborate on complex tasks through delegation, with 100+ built-in tool integrations.

Open-source AI coding extension for VS Code and JetBrains with bring-your-own-model support, fully customizable assistants, and source-controlled AI configuration.

Agent tool integration platform with 1,000+ toolkits, managed authentication, and sandboxed execution connecting agents to Gmail, Slack, GitHub, Linear, and more.

Sourcegraph's AI coding assistant with powerful codebase context. Excels at understanding large codebases through Sourcegraph's code intelligence and provides accurate answers about your code.

OpenAI's open-source AI coding agent that runs in the terminal. Uses OpenAI models to understand codebases, make multi-file edits, and execute commands with configurable autonomy levels from suggestion-only to full auto-approval.

Edge-deployed isolated containers for running AI-generated code, integrated with Cloudflare Workers AI and distributed across 300+ global locations.

Anthropic's official CLI tool for agentic coding with Claude. Enables autonomous code generation, file editing, and command execution directly from the terminal with full context awareness.

Lightweight, developer-friendly embedding database that runs in-memory or persistent mode with minimal setup for RAG prototyping and smaller deployments.

AI observability platform combining evaluation, tracing, and prompt management with CI/CD deployment blocking and cost-quality correlation.

Breakthrough Method of Agile AI-Driven Development. A two-phase framework that orchestrates specialized AI agents — Analyst, PM, Architect, Scrum Master, Developer, QA — through structured planning and implementation workflows with zero context loss.

Git-native issue tracking system by Steve Yegge. Stores work state as structured data in git, enabling persistent agent identity and task management that survives session restarts. Core infrastructure for Gas Town's multi-agent orchestration.

Microsoft's managed agent service with 1,000+ pre-built tools, open-source Agent Framework, and one-click publishing to Microsoft 365 Copilot and Teams.

AWS managed platform for building AI agents that orchestrate multi-step tasks using foundation models, with AgentCore providing runtime, memory, identity, policy, and observability.

Microsoft's multi-agent conversation framework enabling agents to interact, use tools, and collaborate with humans, now merging into the unified Microsoft Agent Framework.

AI software development platform with a semantic Context Engine that indexes entire codebases — code, dependencies, architecture, and commit history — to provide deep context to AI models. Available as IDE extensions, CLI, AI code review, and as an MCP server for third-party agents.

Open-source AI observability and evaluation tool built on OpenTelemetry for tracing, debugging, and scoring LLM applications.

Research updates and technical insights from Anthropic, the makers of Claude. Covers interpretability, safety, and constitutional AI with a focus on responsible AI development.

Anthropic's comprehensive industry report identifying 8 key trends reshaping software engineering — from the shift toward agent orchestration to multi-agent coordination, expansion beyond engineering teams, and security implications.

Deep technical content from the former Director of AI at Tesla and founding member of OpenAI. Features comprehensive tutorials on neural networks, transformers, and AI fundamentals.

AWS AI coding assistant with inline suggestions, agentic capabilities for feature implementation and refactoring, and CLI natural-language-to-bash translation.

Terminal-based AI pair programming tool that works directly with git repositories, creating repository maps for codebase understanding and auto-committing changes.

Comprehensive weekly roundup of AI research, tools, and industry developments. Includes paper summaries, tool reviews, and thought pieces on AI trends.

Community hub for AI-native development featuring 155+ articles, a podcast with 94 episodes, a weekly newsletter, annual DevCon conference, and a comprehensive landscape directory of 624+ AI developer tools.

Practical tutorials and demos of the latest AI tools and frameworks. Focuses on hands-on implementation with clear explanations for developers building AI applications.

Weekly conversations about building AI-powered applications in production. Covers topics like RAG, agents, evaluation, and the practical challenges of deploying LLMs at scale.

Daily AI news analysis podcast hosted by Nathaniel Whittemore (NLW). Covers the full spectrum of artificial intelligence — from new tools and industry disruptions to alignment, ethics, and the path toward general intelligence — six days a week.

Deep technical analysis of LLM research and machine learning trends by Sebastian Raschka, PhD. Covers training methods, fine-tuning, model architectures, and practical implementations with 166k+ subscribers.