Skip to content
Qzentra

Work

Selected systems we’ve engineered.

Case files rather than a portfolio: the problem, the architecture, the decisions, and what was achieved.

Featured case fileAI Systems & Automation

Multimodal AI Sales Funnel

One system that passes through five of the six layers: messaging APIs, multimodal AI, retrieval over real product data, automation logic, and a human handoff.

A crypto-mining hardware enterprise needed inbound social leads qualified, answered with accurate product knowledge, and routed to sales without manual triage.

One production workflow combining messaging channels, multimodal AI, product knowledge retrieval, lead scoring, human handoff, and automated follow-up.

  • n8n
  • WhatsApp
  • Instagram
  • RAG
  • Airtable
  • multimodal
Read the case file
Multimodal AI Sales Funnel: the shape of the system.Two messaging channels and untidy multimodal inputs pass a qualification gate into one engine, which answers from the product catalog beneath it. Scored leads rise to a person; the rest loop back into timed follow-up.

Verified fact500+ASIC miners in the product catalog the system answers from

Layer coverage

  1. Data
  2. Infrastructure
  3. APIs
  4. AI
  5. Automation
  6. Application

Enters throughAI Systems & Automation

Case files

Four more case files, each entering through one service pillar and reaching into others.

  1. AI Systems & Automation

    AI Website Audit & Outreach Engine

    Cold outreach needed to be specific to each prospect, grounded in a real technical audit of their website, and followed up automatically.

    An automated pipeline from target URL to personalized, segmented outreach with follow-ups, built on crawling, PageSpeed analysis, and LLM generation.

    Every message grounded in a Google PageSpeed audit of the prospect’s site

    • web extraction
    • PageSpeed API
    • OpenAI
    • Gemini
    • segmentation
    Read the case file

  2. Data Engineering

    5 TB Zero-Downtime MySQL Migration

    A 24/7 production system on legacy MySQL 5.7 with roughly 5 TB of data had to move to a modern MySQL environment without any service interruption.

    Replication-based migration with compatibility auditing, replica promotion at cutover, and post-migration query modernization. Zero service interruption at cutover.

    5 TBof production data moved with zero service interruption

    • MySQL
    • replication
    • migration
    • SQL optimization
    • high availability
    Read the case file

  3. Data Engineering

    Enterprise Data Query Architecture

    Analytical workloads needed to query and process data across modern data platforms and data lakes without centralizing everything first.

    Query architectures using distributed engines over data lake storage, designed for analytical processing and performance-aware access.

    • Trino
    • DuckDB
    • Apache Drill
    • Azure Data Lake
    • Python
    • SQL
    Read the case file

  4. AI Agents & Voice AI

    AI Voice Receptionist and Outbound Sales Agents

    Inbound calls needed answering and appointments booking without a receptionist on the line, and outbound sales calls needed making without a person dialling every prospect.

    An AI receptionist that books appointments into Google Calendar and outbound AI calling agents for sales, both answering from a knowledge base and handing urgent calls it cannot answer to a person.

    Appointments booked into Google Calendar during the call

    • Voice AI
    • Google Calendar
    • knowledge base
    • LLM APIs
    • telephony
    Read the case file

Engineering range

Each project marked on the layers it worked in, foundation first. Two are data and infrastructure work with no AI in them; that is the point.

  1. Data
  2. Infrastructure
  3. APIs
  4. AI
  5. Automation
  6. Application

A lit mark is a case file that worked in that layer.

Discuss a project

Have a system like one of these, or one that does not exist yet?

Describe the problem and the systems around it. If the answer is a workflow, a pipeline, a migration, an agent, or an application, that is the work we do.

Discuss a Project