Selected work

Relevant delivery experience across AI, workflow design, and operations-heavy systems.

These summaries are designed to show the kinds of environments, workflow problems, and implementation constraints Herod AI is comfortable operating inside.

Book an AI Automation Audit Learn about Mark Herod

AI workflow design
Real-time support systems
Operational communications
Enterprise delivery

Experience

Examples of work that inform Herod AI engagements.

Every project below reflects the same pattern: identify the workflow, reduce friction, improve context, and design the system so it can actually survive production realities.

AI-driven wellness platform

Operational wellness workflows designed around context, follow-up, and visibility

Designed an AI-driven wellness workflow around automated check-ins, transcripts, summaries, alerts, and caregiver or family visibility.

Challenge: The workflow needed to capture context from conversations, reduce manual coordination work, and surface actionable information to the right people quickly.

Approach: Structured the experience around controlled automation, context capture, summaries, and alerts rather than a novelty-first AI interface.

Workflow: Automated check-ins, transcript capture, summaries, alerts, caregiver or family visibility, and context-aware follow-up.

Business value: Improved operational usability, faster context sharing, and better coordination across communication-heavy care workflows.

Security note: Care-related workflows require explicit privacy controls, role-aware visibility, and human review for sensitive communication.

AI summariesWorkflow orchestrationReportingNotifications

Real-time AI support experience

Multimodal support flows built for real-time assistance and production hardening

Designed and prototyped a real-time support workflow combining screen sharing, multimodal analysis, voice guidance, and session intelligence.

Challenge: The system needed to reduce support friction while keeping the user experience fast, understandable, and technically reliable.

Approach: Combined live context capture, spoken guidance, visual analysis, and session-level intelligence into a single operator-focused support flow.

Workflow: React front end, Python services, multimodal vision analysis, spoken guidance, WebRTC, WebSockets, session logging, and Dockerized deployment patterns.

Business value: Created a higher-trust support experience with better context, faster assistance, and a clearer path toward production deployment.

Security note: Real-time support systems need controlled session handling, careful logging choices, and production-minded deployment boundaries.

ReactPythonWebRTCWebSocketsMultimodal AI

AI-enabled communications workflows

Voice and telephony-oriented automation designed around follow-up and reporting

Designed voice and telephony-oriented AI workflows for conversational interactions, reporting, follow-up automation, and communications orchestration.

Challenge: Communication-heavy systems often create fragmented context, inconsistent follow-up, and weak reporting when built as isolated channels.

Approach: Focused on how conversation data should flow into summaries, reporting, tasking, and downstream customer or operator actions.

Workflow: Conversational AI, text-to-speech style interactions, reporting, follow-up automation, and communications orchestration.

Business value: Improved consistency across communication workflows while creating more usable operational context for teams.

Security note: Voice and communications workflows still require access control, logging standards, and safeguards around customer-facing outputs.

Voice AIAutomationReportingFollow-up workflows

Enterprise solution delivery

Operational and commercial improvement inside enterprise IoT and solution environments

Supported enterprise clients, dealers, and technical stakeholders across discovery, solution design, implementation support, and process improvements tied to measurable outcomes.

Challenge: Enterprise delivery environments often struggle when technical complexity, commercial expectations, and cross-team execution are not aligned.

Approach: Combined technical-commercial support, workflow discipline, and solution delivery support to improve execution quality and business outcomes.

Workflow: Discovery, solution design, implementation support, process improvements, and measurable commercial follow-through.

Business value: Contributed to 9% year-over-year IoT growth, improved gross margin from 5% to 10%, and delivered $600K in personal sales while supporting wider team targets.

Security note: Enterprise solution work depends on repeatable governance, documentation discipline, and operational trust across stakeholders.

IoTEnterprise deliverySolution architectureOperational improvement

Book an AI Automation Audit

Want to discuss whether your workflow is a good candidate for AI implementation?

Book an audit to review the workflow, the business case, the data boundary, and the safest first version to test.

Book an AI Automation Audit Review more examples