Technical work / Current practice

Hands-on.
At every layer.

I build and operate the technology I want to understand: private infrastructure, local AI, agent integrations, and connected business workflows.

Private-lab exploration and client advisory work are identified separately below.

01 / Explore the technical layers

From the host
to the workflow.

Each layer creates different questions about access, storage, integration, and control. Select a layer to explore the work.

PRIVATE LAB / CONCEPTUAL SYSTEM MAP

03AI systems & agent orchestrationModel routing · Tool use · Validation
02Services & orchestrationDocker · Python · Bash · Webhooks
01Compute, storage & accessProxmox VE · Linux · ZFS · VLANs

A map of capabilities and dependencies, not a live monitor or an exact deployment topology. No private network details are exposed.

03 / Models & integrations

Local where useful.
Connected where needed.

AI systems span local and hosted inference, agent orchestration, model routing, custom tooling, and validation. Local Whisper transcripts feed downstream workflows; data and trust boundaries remain explicit.

Explore the AI work

02 / Services & orchestration

Turn individual tools
into a working process.

Docker services, Python-based custom skills, Bash automation, webhooks, and REST API integration testing provide the connective work between systems.

Explore business automation

01 / Infrastructure

Understand what
the software relies on.

Proxmox virtualization, Docker containerization, ZFS storage, Linux administration, network segmentation, and identity and access controls keep the infrastructure work hands-on.

Explore the infrastructure

02 / Private lab · Infrastructure

A practical foundation
for technical judgment.

The lab gives me direct experience with the layers underneath an application: hypervisors, containers, storage, permissions, and networks. It keeps conversations about infrastructure grounded in operational decisions.

I provision and manage Proxmox-based environments, Docker services, and ZFS storage, alongside Linux administration, VLAN segmentation, and access controls.

Scope: personal infrastructure & R&D

Infrastructure / Dependency view

Applications & servicesAutomation, local AI, and containerized workloads

VirtualizationProxmox VE
Compute & workload isolation

ContainersDocker
Service environments

Storage & OSZFS · Linux
Data & administration

Network & accessVLANs · IAM
Segmentation & permissions

Conceptual dependency view of the technologies used in my lab. The diagram groups responsibilities; it does not imply every service uses every layer.

The decisions behind the diagram

Where does responsibility sit?

A service depends on more than its application settings. Storage access, workload isolation, and network boundaries all shape whether it can operate as intended.

What carries into a buyer conversation?

A concrete way to discuss integration constraints, access requirements, and ownership with infrastructure and security teams.

03 / Private lab · AI systems engineering

Build with models.
Design the system around them.

Agent orchestration · Integration · Business-process modeling

I build and experiment with AI systems that combine local inference, hosted models, agent orchestration, automation, and custom tooling.

The work is in the decisions around the model: how a task is decomposed, which model handles it, what context and tools it receives, how agents coordinate, and how the result is validated.

Architecture explorer

Follow the task. See the boundary.

Lab patterns / Conceptual view

Local audio path

AudioWhisper ASRTranscript

Transcribed locally, then passed into a downstream workflow.

01
Input, task & context
A request or transcript enters the workflow.
02
Agent orchestration & custom tooling
Task decomposition · Context management · Agent coordination

CodexOpenClawPython

Tool use & integrationsMCP · REST APIs · Webhooks
03
Model selection & routing
Match the task to a model and an allowed endpoint.

LiteLLMOpenRouterLocal / hosted endpoints
↙ Local inferenceOllama / Mistral
Self-hosted model pipelines
Within the local inference boundary

↘ Hosted inferenceOpenAI / Anthropic
External model services
Crosses an external service boundary
04
Validation & fallback paths
Check the result against the task. Revisit context, tools, or model selection when needed.
↓ Result to downstream workflow

Patterns built and explored in my lab, not one monolithic production AI application. A routing layer connects options; it does not erase their boundaries.

Path 01 / Local inference

Keep inference
on local infrastructure.

Ollama and Mistral provide self-hosted model pipelines. The system still needs task decomposition, context management, tool permissions, and validation.

Data & trust: Inference stays local. Tool calls need their own boundary review.
Cost: Local compute and model capacity.
Operations: Local endpoints, resources, and model availability.

Path 02 / Hosted inference

Cross the boundary
deliberately.

Model routing through OpenRouter, LiteLLM, and hosted endpoints connects external services such as OpenAI and Anthropic to agent workflows.

Data & trust: Selected context leaves the local inference environment.
Cost: Provider usage, model selection, and call budgets.
Operations: Service availability, API behavior, and fallback choices.

Path 03 / Audio into the workflow

Start with audio.
Route the transcript.

Whisper turns audio into text on local infrastructure. That transcript becomes input and context for downstream agent, tooling, or business-process workflows.

Data & trust: Local transcription does not make every later step local.
Next decision: Choose the model, tools, and context the task requires.
Validation: Check transcription and downstream results against the task.

The boundary is part of the design. Local inference and hosted services have different data, security, cost, and operational boundaries. The model is one component; orchestration, context, tool access, routing, validation, and fallback paths determine how the wider system behaves.

04 / Client advisory · 2024–Present

Modernize the process,
not just the toolset.

For an established field-services business, I built and integrated an operating stack spanning CRM routing, automated pricing, programmatic invoicing, and customer payment gateways.

The work also included automated inbound lead pipelines, custom email domains, API-driven customer communications, and targeted digital marketing.

Scope: implemented business & operational modernization

Field services / Connected operating flow

Capture & route the enquiry
Inbound lead pipelines and custom CRM routing.
Price & communicate
Dynamic automated pricing and API-driven customer communications.
Invoice & collect payment
Programmatic invoicing and instant-deposit payment gateways.

A simplified view of the business functions connected by this work. Client identity, operational details, and credentials remain private.

What changed, and what is documented

A legacy business gained connected workflows across acquisition, pricing, invoicing, and payment. The documented evidence is the implementation scope; no time-saved, conversion, or revenue metric is claimed here.

The connection to enterprise work

Better questions.
Earlier in the evaluation.

Hands-on work helps me enter architecture and security discussions with context, identify the real constraint, and work effectively with engineering. The commercial history shows where that technical fluency meets procurement, executive alignment, and revenue.

Explore the commercial evidence

Technology with a purpose

What needs to work better?

Discuss the problem

From the hostto the workflow.

Local where useful.Connected where needed.

Turn individual toolsinto a working process.

Understand whatthe software relies on.

A practical foundationfor technical judgment.