Full-Stack + AI / Computer Vision

I ship AI products, automations, and the infrastructure that keeps them reliable.

I build end-to-end systems: computer vision pipelines, agent workflows, dashboards, and deployments with logs, retries, and operational clarity.

Contact / Book a Call View Case Studies

AI PipelinesDashboardsMCP/DevOpsMobile + Web

Jeruzael

Builder portfolio

UTC+8

10+

Apps shipped

AI/CV

Pipelines

Ops-ready

Automations

Toolchain

PythonFastAPIFlutterDockerYOLO/SAMn8nMCP

Signal

Systems That Ship

I bias toward practical reliability: clear boundaries, server-side validation, and checks that prevent silent failures.

PythonTypeScriptFastAPINext.jsFlutterDockerPostgresPlaywrightn8nMCP

Background: DOST-FNRI experience building applied AI tools, plus automation systems that include monitoring, retries, and operational guardrails.

Featured

Case Studies

Deep technical proof: architecture, implementation constraints, and outcomes.

View all

AINA (AI Nutrition Assistant)

Mobile AI assistant with a clear data flow from capture to inference to actionable results.

Details

Problem: Users need nutrition insights that feel immediate and usable, not buried in spreadsheets.

Approach: Define an inference boundary, keep UX fast, and make model outputs explainable enough to trust.

Result: A product-shaped experience with an architecture that supports iteration on models without rewriting the app.

FlutterPythonFastAPICV/MLStorageTelemetry

AI Agent Business Automation + Dashboard

Orchestrated workflows with audit logs and a dashboard that surfaces state and failures.

Details

Problem: Business workflows fail silently: missed leads, inconsistent handoffs, and no visibility into why.

Approach: Treat workflows as products: retries, idempotency, logging, and a control surface for humans.

Result: Less manual chasing, fewer dropped tasks, and a dashboard that makes operations legible.

n8nTypeScriptPythonAPIsDashboardObservability

MCP Server + Local Codex/n8n Tooling

A tool registry boundary that lets agents call real capabilities safely and consistently.

Details

Problem: Agents become unreliable when tool access is ad-hoc, insecure, or inconsistent across environments.

Approach: Centralize tool contracts, keep auth boundaries explicit, and document what tools do and do not do.

Result: More reliable agent execution and fewer ambiguous failures during automation runs.

MCPNode.jsPythonAuthToolingAutomation

Computer Vision Training Pipeline (YOLO/SAM)

Dataset to training to evaluation with reproducibility and deployment readiness.

Details

Problem: CV work fails when data, training, and evaluation are not repeatable or measurable.

Approach: Standardize dataset flow, version configs, and enforce evaluation outputs as artifacts.

Result: A pipeline that supports iteration without losing track of what changed and why.

YOLOSAMPythonData pipelinesExperiment tracking

Capabilities

Systems I Build

Four buckets that cover most real work.

AI / Computer Vision Systems

Training, evaluation, deployment readiness, and the glue that makes models usable in products.

Dataset + annotation flow
Reproducible training configs
Inference services + monitoring

Automation and Agent Workflows

Orchestrated pipelines that are observable, retry-safe, and auditable.

Tool routing + context providers
Retries + idempotency
Human-in-the-loop control points

Dashboards and Data Pipelines

Interfaces and pipelines that let humans monitor and steer complex systems.

Operational dashboards
Event logs + audit trails
Metrics and failure surfacing

Secure Deployment / DevOps

Deployments with clear rollback paths and environment hygiene.

Docker + CI checks
Secrets discipline
Release verification checklists

Trust

Architecture Highlights

Diagram-friendly containers now, real diagrams later. The goal is to make the system legible.

Architecture

AINA: Mobile + Inference

Diagram placeholder (16:9)

Key components

Mobile app
Inference service
Storage + telemetry
Model lifecycle

Architecture

Agent Automation: Tools + Orchestration

Diagram placeholder (16:9)

Key components

Workflow runner
Tool registry
State + retries
Dashboard + audit log

Architecture

CV Training Pipeline

Diagram placeholder (16:9)

Key components

Dataset pipeline
Training jobs
Evaluation
Packaging for deployment

Breadth

Projects

A wider set of work, with consistent structure and proof links.

Ops Dashboard Patterns

Reusable UI patterns for stateful workflow monitoring: queue status, retries, and audit trails.

DashboardsUXObservability

Model Evaluation Harness

A repeatable evaluation runner that produces artifacts and comparable runs.

CVMetricsReproducibility

Deployment Baselines

Build + deploy patterns with rollback notes and release verification checklists.

DevOpsDockerCI

Tool Registry Concepts

Practical boundaries for tool contracts, auth scopes, and safe tool execution.

MCPSecurityAutomation

CV Data Labeling Flow

A workflow to keep annotation quality measurable and reviewable.

DataCVWorkflow

Client Intake Templates

High-signal templates to capture constraints, success metrics, and operational requirements early.

DeliveryProcessPM

Experience

Timeline

Skimmable evolution, kept intentionally adult.

DOST-FNRI
Applied AI / CV work
Shipped practical AI tools with real constraints: data quality, reliability, and usable UX.
AINA
Mobile AI nutrition assistant
End-to-end product: mobile UX, model integration, and system design choices made explicit.
Automation
Agent workflows + dashboards
Orchestrated workflows with auditability, retries, and observable state.
Pipeline
CV training and evaluation loop
Dataset -> training -> metrics -> packaging, with reproducibility and deployment readiness.

For Clients

Services

Two delivery tracks depending on what you need shipped.

Product Build (Web/Mobile + AI)

Ship a usable product: UX, API, model integration, and deployment.

MVP scope mapping
Model integration
Operational baseline (logs, retries)

Typical: 2-8 weeks

Start a conversation

Automation Build (Agents + Dashboard)

Automate workflows with visibility, auditability, and control surfaces.

Workflow orchestration
Tooling + integrations
Monitoring + audit log dashboard

Typical: 1-6 weeks

Start a conversation

Conversion

Contact

Tell me what you want to automate or build. I will map the system and the fastest path to shipping.

Availability

Open to full-time roles and scoped client builds. If you have a messy workflow, bring it.

jeruzael.d@gmail.com

Links

LinkedIn GitHub

Message Template

If you want a fast, high-signal reply, send something like this:

Subject: Build / automate <X>

Goal:
Constraints:
Deadline:
Current stack:
Data sources / integrations:
What "done" looks like: