The Real Python Podcast – Episode #304: Setting Up a Flexible LLM Harness and Web Scraping with Scrapy
This episode examines whether model or harness matters more, plus agentic workflows, web scraping, and self-hosting for reliable Python AI tools.
AI Agents in Practice
AI agents promise to do multi-step work on their own. This topic looks at what actually works today: real automations, no-code agent builders, coding agents and the guardrails you need before letting an agent touch your email, files or spreadsheets.
Use Grok 4.7 on Amazon Bedrock to create long-horizon agents
Grok 4.7 on Amazon Bedrock helps build long-horizon agents; choose reasoning effort to balance accuracy and token cost, with Bedrock profiles, APIs.
Where the Agent Development Lifecycle Belongs
Agent lifecycle should align with app delivery through shared checkpoints, owners, tests, and monitoring for safe AI features.
OpenAI's cutting-edge models and Codex are now accessible on AWS
OpenAI's frontier models and Codex are now generally available on AWS, allowing enterprises to access these AI tools directly within their existing cloud ecosystem. This eliminates friction from separ…
Run fast and private on-device agents with LFM2.5-24B-A2B.
LFM2.5-24B-A2B is a local reasoning agent with two billion active parameters, fast Apple M5 Max inference, and a 131,000-token context.
Tokens, Context Windows, and Caching: What You’ve Been Getting Wrong
Claude agents fail on cost and latency from token fragmentation, tool schemas, context growth, and decode. Use count_tokens, caching, short answers.
AI-Powered Python Coding Learning Roadmap
Use AI assistants like Cursor or Claude Code in a daily Python workflow to draft, debug, test, and review code while keeping judgment central.
Developing agents that conduct transactions with Anchor Browser and AgentCore payments
Anchor Browser and AgentCore payments automate HTTP 402 purchases for AI agents with budgeted sessions, managed wallets, retries, and audit logs.
Enhancing computer use through Ironclad
OpenAI and Ironclad test AI agents for contract workflows, exploring computer use for professional tasks, pending safety, cost, and reliability.
The Invisible API Revolution: How AI Agents Quietly Reshaped the Internet’s Plumbing
AI agents are becoming primary API consumers. MCP, A2A, and agentic protocols may change API design, traffic patterns, rate limits, and value metrics.
Supplementary Note to the GPT-5 System Card: GPT-5-Codex
GPT-5-Codex is OpenAI's specialized version of GPT-5, optimized for agentic coding within the Codex environment. Announced via a system card addendum, its standout feature is dynamic thinking effort: …
How Codex helps Endava create an agent-driven organization
Global technology services firm Endava is transforming into an 'agentic organization' using OpenAI's Codex. By integrating AI agents across the software delivery lifecycle—from requirements analysis t…
Learn How to Rewrite Apache Spark User-Defined Functions for GPU Acceleration with AI Agents
NVIDIA cuDF skills use AI agents to rewrite Spark user-defined functions for GPU acceleration, avoiding CPU fallbacks and boosting jobs by 22x+.
Citing @joedaroo
Sudden AI capability jumps can expose weak processes. Organizations need trained teams, clear escalation, and flexible controls to secure agents.
Inside Claude Code: Understanding How Context, Tokens, Tools, and Subagents Work
Claude Code assembles layered context from rules, Skills, MCP, files, and tool output, then manages tokens and subagents while running tests.
Use Amazon Bedrock AgentCore to build document processing that improves itself
Amazon Bedrock AgentCore supports agentic document processing that self-learns invoice formats, reducing manual work and maintenance costs.
The Rise of AI Agents: From AutoGPT to Claude Code
AI coding agents have moved beyond AutoGPT-style autonomous loops toward narrower, workflow-aware tools like Claude Code, Codex, Cowork, and OpenClaw. The key insight is that agent value now depends o…
Are You Considered a Codex Original?
Codex Originals is collecting real stories from builders, tinkerers, researchers, and creators who use Codex. The notice is an invitation, not a formal contest, so applicants should share practical pr…
Introducing gpt-realtime along with updates to the Realtime API
OpenAI announces gpt-realtime, a more advanced speech-to-speech model enabling faster, more natural voice conversations, alongside major Realtime API updates: MCP server support for connecting externa…
How Codex helps Virgin Atlantic accelerate its shipping
Virgin Atlantic's engineering team faced an immovable deadline: revamp its mobile app before the holiday travel rush. By integrating OpenAI's Codex into daily workflows—generating unit tests at scale,…