Tag
ZipHQ rebuilt their agent on LangGraph + LangSmith, cutting development time for new features and gaining automatic tracing for LLM calls, storage, and context without writing code.
The Philadelphia Inquirer built Scrape, an in-house AI tool developed with the Lenfest Institute, to surface hyperlocal news leads from fragmented local sources such as municipal meetings and school calendars, cutting down on the ~12 hours per week an editor spent finding newsletter items through iterative prompt refinement and embedded editorial context.
OpenAI case study on how Chatham Financial uses Codex and GPT-5.6 to power internal tools, its Chatham Onyx platform, and its Process Zero consulting service — cutting trade validation from 30 minutes to under 4 minutes. The work spans employee-built apps via Chatham Vibes and client-facing capital markets workflows.
OpenAI shares a customer story about The Den Family Social, a Denver-based social club, using ChatGPT Work to save 10-15 hours a week on administrative tasks like licensing and grant applications while expanding to a second location.
A Canadian real-estate agent with 60k+ stale CRM leads got a text and voice AI system built for $5k, which re-engaged old leads, flagged ready buyers, and produced 34 closed deals worth ~$490k in commission over seven months. The post details outreach numbers, consent handling, follow-up scheduling, and lessons learned.
Vercel highlights the migration of the Mercedes-AMG PETRONAS F1 Team website, achieving 70% faster builds and 75% faster paints with minimal intervention, and derived two AI skills to be shared.
Proaction, a software company for fleet management, uses Codex to create personalized demos, boosting sales by 60% and saving over 75 hours monthly in engineering work.
Marina Pasqual describes how setting explicit boundaries for AI agents enhances their utility in operational tasks like community growth, ensuring human judgment is retained for critical decisions.
The article argues that AI's true power lies in mass collaboration through multi-agent systems, exemplified by OpenAI's swarm architecture solving the Navier-Stokes problem rapidly.
IBM's open-source agent CUGA integrates Tavily for web search, enabling reliable access to live information in air-gapped and sovereign environments, and streamlining enterprise applications.
The article details a pilot project where a team tested client data for a RAG system, initially achieving only 40% accuracy due to issues like conflicting documents and poor retrieval. Through routing, data cleaning, and acronym mapping, accuracy improved to about 90%, highlighting the critical role of data readiness in successful AI transformation.
Shopify built a continual learning loop using PyTorch and vLLM to improve their GraphQL agent, reducing costs by 96% and outperforming frontier models through production-driven updates.
The blog post explores the effectiveness of generative testing versus unit tests in discovering bugs, using the Rust regex crate as a case study. It demonstrates how a custom fuzzer found multiple bugs and provides techniques for improving testing approaches.
Included Health built Dot, an AI-powered federated multi-agent healthcare guide, using LangGraph and Deep Agents, resulting in a 75% increase in chat engagement and over 99% high-risk detection.
LangChain demonstrates how LangGraph and LangSmith have accelerated support agent development at Lyft, reducing shipping time from six months to 1-2 weeks with enhanced visibility.
Rox, an AI revenue agent company, improved its agents' performance by switching to Tavily's web search API, reducing sales research time from weeks to seconds while enhancing data freshness and lowering costs.
An AI agency shares their experience implementing AI in a client's support system, detailing benefits like improved self-service and agent efficiency, but noting that AI cannot fix underlying policy issues and requires ongoing maintenance.
BotsCrew improved team handoffs and context sharing by using Claude AI as a shared resource with structured Instructions, Context, and Playbooks, leading to faster bug reporting and efficient workflows.
This paper evaluates the reusability of publicly available speech resources for low-resource languages, focusing on Central Kurdish as a case study.
OpenAI published a case study called 'Defense Factory' describing how it used Codex agents and cyber models in a company-wide security sprint to find and fix vulnerabilities across hundreds of systems, sharing a practical playbook for building continuous agent-based security loops.