Home on Veratai

A Deep Research Agent for Underwriting

Mon, 01 Jan 0001 00:00:00 +0000

By developing a deep research agent, we allowed an international insurer to dramatically reduce the time to search for and collate risks — leaving underwriters to focus on analysis and shortening time-to-bind.

Clinical terminology translation for SNOMED International

Mon, 01 Jan 0001 00:00:00 +0000

In just 11 days, we helped SNOMED International prove that an AI agent could accurately translate precise clinical terms into native languages — paving the way to increase adoption worldwide.

How to design a Deep Research Agent (that’ll actually work)

Thu, 09 Apr 2026 13:59:14 +0000

Here’s an unfashionable opinion in 2026: most of the work of building a good agent has nothing to do with the agent. It’s data curation, eval design, workflow mapping, and arguing about schemas. The coding bit is maybe 20% of the task. This blog covers a good chunk of the remaining 80%.

How to create a Self-improving AI agent

Fri, 23 Jan 2026 09:54:32 +0000

If you’re developing an AI agent and you want it to improve itself with usage, this post is for you.

How to Kick Off a Deep Research Agent Build

Tue, 30 Dec 2025 10:01:50 +0000

Our experience shows that the most successful Agentic AI projects begin with specific preparation steps. In this blog, the first in our series on how to build AI agents, we’re going to take a look at Veratai’s process for designing deep research agents for the enterprise.

How to build a research agent for your team

Thu, 04 Dec 2025 09:03:04 +0000

At Veratai, we’ve been building AI agents for organisations that want to automate complex information gathering workflows. Here is our five-phase methodology for reliably delivering working systems rather than expensive proof-of-concept graveyards.

DIY Deep Research Agents

Tue, 21 Oct 2025 08:58:07 +0000

“Deep Research” is one of the breakout agentic use cases of 2025. Having run well over a hundred deep research tasks and built three custom agents for clients, here are the eight factors that decide when “rolling your own” research agent is the superior choice.

Why is AI like Interstellar Travel?

Fri, 10 Oct 2025 09:18:40 +0000

The Incessant Obsolescence Postulate says that any spacecraft you launch today risks being overtaken by a future one. AI adoption faces the same dilemma: deploy now, or wait for better models? Here’s why, in most cases, you should launch now.

Pick the right problem: How to Avoid AI Project Failure

Wed, 24 Sep 2025 07:00:00 +0000

Artificial Intelligence is no longer a sci-fi dream — it’s woven into the fabric of modern business. But here’s the catch: as many as 95% of generative AI projects fail, especially in sales and marketing. So how do you ensure yours lands in the 5% that succeed?

Picking your agentic AI toolset: a guide for the perplexed

Thu, 04 Sep 2025 14:42:00 +0000

In this essay I’ll compare three real-world agentic AI platforms - n8n, Lindy and CrewAI. They neatly represent different points on the “agentic” spectrum: the agent-on-rails (n8n), specialised (“assistant-like”) agents (Lindy) and the high autonomy, multi-agent toolkit (CrewAI).

How to find your best AI Automation opportunities

Mon, 01 Sep 2025 14:45:00 +0000

In this post, we introduce a simple, 2x2 matrix that we use to triage and organise AI automation opportunities. It will help you to determine which ideas to pursue and how to get going.

4 ways companies that value their customers should use AI (and 4 to avoid!)

Tue, 14 Jan 2025 11:51:34 +0000

Consumers are becoming wise to “AI slop”. For companies that put the customer experience first, here are four ways we recommend using AI – and four to avoid.

Podcast: Deep Dive into Vision Language Models (VLMs)

Tue, 07 Jan 2025 14:22:03 +0000

Over Christmas, Nathan Labenz and our Data Science Director Will Hardman spent 4 hours doing an extended deep dive into Vision Language Models (VLMs) on Nathan’s awesome podcast, The Cognitive Revolution. The podcast can be found on The Cognitive Revolution website, or wherever you get your podcasts.

Developing an AI Strategy: Analysis & Synthesis

Mon, 09 Dec 2024 11:50:23 +0000

Analysis & Synthesis is the third (and final) stage of Veratai’s methodology for developing AI strategies. How can you take a list of ideas and turn them into a winning strategy? We explain the alchemy of the process in this blog.

#The_Deep_End: Alignment post-training for LLMs: lessons learned making it work

Mon, 25 Nov 2024 14:05:59 +0000

Alignment post-training is the art of getting an AI model to complete tasks in a way that meets the criteria of its users. In this blog, I document what I learned getting this process to work for a production-grade LLM.

Developing an AI Strategy: Effective Research

Mon, 11 Nov 2024 15:45:05 +0000

The second post in our series on developing an AI strategy covers Research & Ideation: how to build background knowledge, ask the right questions, find and evaluate good sources, and develop a long-list of big ideas.

The first step toward an AI strategy: good stakeholder interviews

Wed, 09 Oct 2024 15:13:00 +0000

Good stakeholder interviews are the first - and perhaps the most critical - step in developing an Enterprise AI Strategy. Learn how to do them successfully in this blog post.

Book a 30-minute Consultation

Mon, 01 Jan 0001 00:00:00 +0000

We bring our expertise to help you understand how your team could benefit from agentic AI and how you can overcome the challenges of getting agents widely used. We end with actionable next steps so you can get started.

What to expect

A relaxed, 30-minute conversation
Guidance on where agentic AI can (and can’t) help your organisation
Practical tips on the “trip-wires” ahead and how to overcome them
Clear, practical next steps you can act on straight away

Rather email?

No problem. We would love to support you in your AI journey. Please email us at hello@veratai.co.uk and we’ll get straight back to you.

Our Leadership Team

Mon, 01 Jan 0001 00:00:00 +0000