/news4 min read
What We Shipped: July 2026
6 use cases, ISO 27001, new integrations, and a world tour.
Rotem Weiss named to Forbes 30 Under 30
Tavily Founder and CEO, Rotem Weiss, was named to the Forbes 30 Under 30 list this month. He leads with the same clarity and focus he expects from the product and team - no shortcuts, no noise, just building things that matter. Congratulations Rotem!
Tavily is now ISO 27001 certified
Security has always been a core part of how we build. Earning the ISO 27001 certification makes that commitment official.
ISO 27001 is the international standard for information security management, covering how we handle data, manage risk, and protect the infrastructure your agents depend on. For teams building in regulated industries or evaluating enterprise deployments, this gives you a documented, independently verified baseline to work from.
Review our security posture, compliance documentation, and certifications.
On the road: Seoul, Berlin, and Paris
July was a busy month on the ground. The Tavily team traveled to three major events: ICML in Seoul, WeAreDevelopers World Congress in Berlin, and RAISE Summit in Paris. We enjoyed connecting with researchers, builders, and engineering leaders across the AI ecosystem at every event.
If you missed us, we're not slowing down. Find us at an upcoming event near you.
Nebius' Agentic AI Builder certification
Nebius launched a new certification program for software engineers building agentic AI workflows, with Tavily and Token Factory built into the curriculum. The Agentic AI Builder certification validates skills across the inference and tool-calling layers, testing concepts relating to RAG pipelines, post-training / fine-tuning, and agentic search with Tavily.
It's a practical, role-based exam: 20–30 questions, 60 minutes, delivered remotely. During the early-bird period, registration is $1 (regular price: $100).
Tavily is now live in…
Tavily keeps showing up across more of the tools builders are already using. Two new integrations went live this month:
- Nemo Deep Agents: Tavily is available as a managed egress integration for NVIDIA NemoClaw's nemo-deepagents sandbox. If you're running LangChain Deep Agents Code inside a NemoClaw OpenShell sandbox, you can enable Tavily web search and extraction through a tavily policy preset, keeping your API key on the host gateway, outside the sandbox. See the setup docs →
- Grok: Real-time search is available inside Grok Build. Fire up /plugin, add Tavily, and your agents get clean, citable results grounded in live web data, no extra infrastructure required. Try it in Grok Build →
What Tavily powers: 6 production use cases
Tavily shows up across industries and workflows. Here are six use cases teams are running in production today:
- Sales account intelligence: Research prospects and accounts before outreach, grounded in live web data
- Product and competitor intelligence: Track competitor moves, pricing changes, and product announcements in real time
- Academic and scientific research: Pull current papers, preprints, and citations for research workflows
- Investment research briefs: Aggregate financial signals, news, and filings for analyst-grade reports
- Threat intelligence enrichment: Enrich security incidents with live threat feeds and CVE data
- Vendor risk and KYC screening: Surface regulatory, litigation, and reputational signals for due diligence
All six are available as installable agent skills: npx skills add tavily-ai/use-case-skills
Architecture behind production AI agents
Most teams can build an AI agent. Far fewer can deploy one that's reliable, observable, and cost-effective in production.
On August 4, Tavily's Lakshya Agarwal joins speakers from Nebius and LangChain for a live session on the architecture behind production agents. The session uses a real-world compliance audit agent as the case study. It’s built with LangChain Deep Agents, Tavily, and NVIDIA Nemotron 3 Ultra on Nebius Token Factory, and walks through grounding, retrieval, observability, and simulation testing. You'll also see how an open-model architecture compares to a frontier model on cost and quality.
Tuesday, August 4 · 6:00 PM UTC+2 / 9:00 AM PDT · Online
How Fieldway Built a Product Research Agent With Tavily
Matthew Stublefield, CEO of Fieldway, wired Tavily into Claude Code to replace six-figure research firms with his own agent — turning a 5-hour research cycle into 25 minutes, with every citation verified before he sees it. On one project, it surfaced a competitor a bank had never tracked, the actual source of their churn, cutting a 3–4 week engagement down to 10 hours. "I'm now getting something better than what those six-figure firms produced, in 25 minutes to 2 hours."
Check out the full case study →
Loving Tavily? Tell the world
Real reviews from real users carry more weight than anything we could say ourselves. If Tavily has made your work easier, two minutes of your time could be exactly what someone else needs to find the right tool.
Connect with us
Questions or feedback on any of this? We're on Discord, X, and LinkedIn.