Field notes · pragmatic AI
Writing for people who ship AI —
not people who talk about it.
Deployment playbooks, compliance guides, operating-model thinking, and the ROI math we actually use with clients. Written by the team that has the conversations.
The evaluation harness: engineering the layer between your AI and production
Most AI systems don't fail because the model is wrong — they fail because nobody built the harness. The evaluation layer that catches drift, enforces guardrails, and tells you when the model is lying. Here's how to engineer one that survives production.
Read the playbook →More from the archive
Omnichannel Voice AI: What the SoundHound Deal Means
Omnichannel voice AI strategy after SoundHound's $250M LivePerson deal: why UK enterprise buyers should go voice-first now and integrate omnichannel later.
Voice biometric data security: enterprise GDPR obligations
AI voice biometric data triggers GDPR Article 9 — explicit consent and a mandatory DPIA. What UK enterprises must audit before deploying voice automation.
Vertical AI voice agents: build for industry or fail
Vertical AI voice agents enterprise: Avoca's $1B raise proves industry-specific agents outperform generic platforms. Here's what that means for your sector.
Enterprise voice AI vendor evaluation: what buyers ask
Enterprise voice AI vendor evaluation goes beyond demos. Learn what IT, Legal, Finance, and Operations actually scrutinise — and the benchmark answers.
Enterprise AI voice agents: the complete guide
Enterprise AI voice agents reduce per-call costs by 90%. Covers architecture, ROI, and compliance for UK enterprise teams deploying voice AI at scale.
AI voice healthcare appointments: the enterprise guide
AI voice healthcare appointments cut no-shows by up to 81%. 8.1m NHS DNAs cost £1.2bn a year — this enterprise deployment guide covers ROI and ICO compliance.
EU data residency voice AI: enterprise compliance guide
EU data residency for enterprise voice AI is now a deal prerequisite, not just a checklist item. What GDPR, UK ICO, and EU AI Act require before deployment.
AI receptionist enterprise: infrastructure and ROI guide
AI receptionist enterprise: UK contact centres lose £6.7bn to repeat calls. Infrastructure, ROI, and compliance essentials for enterprise voice AI buyers.
AI SDR automation ROI: the enterprise buyer's guide
AI SDR automation ROI: real cost benchmarks, GDPR and PECR compliance requirements, and the internal approval framework enterprise buyers actually need.
EU AI Act and voice AI: enterprise compliance guide
EU AI Act and voice AI: what enterprises must do before 2 August 2026. Article 50, emotion AI obligations, and a compliance checklist for contact centres.
Consent capture in AI voice calls: GDPR and PECR guide
Consent capture in AI voice calls governs your entire outbound programme under GDPR and PECR. Get the lawful basis framework UK enterprises use pre-launch.
Business case AI voice: the enterprise framework
Business case AI voice automation: the complete framework for modelling ROI, implementation cost, risk and payback. Built for enterprise finance teams.
Inbound vs Outbound AI Voice Agents: Practical Guide
Inbound vs outbound AI voice agents: which should enterprises deploy first? A practical decision framework covering use cases, ROI, and deployment sequencing.
AI Voice Cost Per Call: Human, Hybrid, and AI Economics
AI voice cost per call: benchmark human, hybrid, and AI economics. A practical framework for enterprise teams making the business case for voice automation.
Building Dilr Voice on Pipecat: the open-source voice AI framework powering thousands of production calls
Dilr Voice handles thousands of production calls daily on the Pipecat open-source framework. A deep-dive into the architecture: multi-agent routing, real-time telephony, knowledge base RAG, and the six systems we engineered on top.