ADArsenios DiamantakosApplied AI Engineer
Applied AI Engineer

Applied AI systems built to be inspected, run, and trusted.

I build Python, FastAPI, automation, reporting, and ML workflow projects with the proof hiring teams need: clear scope, runnable code, tests, screenshots, and reviewable outputs.

Python/FastAPI implementation work
Data pipelines and reporting outputs
Optional AI layers with deterministic fallbacks
Tests, docs, screenshots, and public-readiness notes
Interactive review mode

Choose how a reviewer enters the portfolio.

The interaction changes the evidence path instead of only changing colors. It is designed for fast hiring review: scan, inspect, verify.

Recruiter scan

Fast path to fit, proof, and contact.

Start with the public repos, check the case-study summaries, then open the CV package. The site keeps private-ready work labeled so there is no overclaiming.

Public GitHub repos
Project pages
CV package
Contact path
Evidence map
7 public / 1 private-ready

Agentic Trading Research Lab

Python / CLI

private

CaseForge Studio

Python / CLI

repo

Job Market Intelligence Pipeline

Python / Typer

repo

AI Workflow Automation Assistant

FastAPI / Pydantic

repo

LinkedIn Ads Daily Check

Python / CSV

repo

ViT Pet Classification Pipeline

PyTorch / Transformers

repo

FlowCore Master

Bash / Oracle SQL

repo

AI Nutritionist

Python / Streamlit

repo

About

Practical AI implementation with evidence a reviewer can actually inspect.

AI and software engineering graduate with a BSc in AI & Computer Science and an MSc in AI & Business Strategy. My work focuses on turning messy operational inputs into tested, documented systems: FastAPI services, Python CLIs, reporting pipelines, local-first assistants, optional OpenAI layers, and ML inference demos with deterministic baselines, public-ready documentation, and clear validation.

Automation
Data/reporting
ML delivery

Projects

A curated library of implementation work.

Public repos, private-ready projects, and planned releases are separated clearly. Hiring reviewers can open the case study first, then jump to GitHub or download a ZIP where the repo is intentionally public.

Public repos

7

ZIP downloads

6

Project pages

8

Private-ready

1

Project explorer

Risk-controlled research system

Agentic Trading Research Lab

Agentic Trading Research Lab static dashboard

Risk-controlled research system

Agentic Trading Research Lab

Private

Private paper-trading research lab with deterministic risk gates, offline fixtures, strategy comparison, walk-forward reports, audit logs, and a static dashboard.

PythonCLIpytestDockerSQLiteBacktesting
Review proof
pytest suite
paper-only safety audit
release smoke script

Risk-controlled research system

Agentic Trading Research Lab

Agentic Trading Research Lab static dashboard

Risk-controlled research system

Agentic Trading Research Lab

Private

Private paper-trading research lab with deterministic risk gates, offline fixtures, strategy comparison, walk-forward reports, audit logs, and a static dashboard.

pytest suite
paper-only safety audit
release smoke script
PythonCLIpytestDockerSQLite

Reporting utility

LinkedIn Ads Daily Check

LinkedIn Ads Daily Check overview report

Reporting utility

LinkedIn Ads Daily Check

Public

Compact Python utility that turns LinkedIn Ads exports into KPI, pacing, campaign-priority, HTML, JSON, and Markdown reports.

unittest suite
Sample report outputs
Public GitHub repo
PythonCSVHTML reportsJSONMarkdown
Local-first workflow automationPublic
Intake
AI/mock
Review
request validated
decision drafted
review required

Local-first workflow automation

AI Workflow Automation Assistant

Public

FastAPI workflow app for operations intake triage, structured AI/mock decisions, SQLite persistence, and review queue preparation.

pytest suite
ruff
GitHub Actions CI
FastAPIPydanticSQLAlchemySQLiteOpenAI optional
Implementation blueprint generatorPublic
Plan
Architect
Evaluate
Implementation dossier

Implementation blueprint generator

CaseForge Studio

Public

Turns rough product ideas into implementation-ready blueprints with a deterministic pipeline, CLI, local web app, saved runs, comparison views, and optional OpenAI refinement.

Unit test suite
Build workflow
Release checklist
PythonCLIHTTP APIlocal web UIunittest

Data pipeline and reporting

Job Market Intelligence Pipeline

Job Market Intelligence Pipeline report desktop screenshot

Data pipeline and reporting

Job Market Intelligence Pipeline

Public

ATS-backed job market pipeline for Greenhouse/Lever data collection, validation, enrichment, DuckDB history, delta reports, and fixture-backed demos.

24 pytest tests
GitHub Actions CI
Package build
PythonTyperDuckDBPandasBeautifulSoup

Computer vision delivery path

ViT Pet Classification Pipeline

ViT Pet Classification Pipeline Streamlit UI without third-party branding

Computer vision delivery path

ViT Pet Classification Pipeline

Public

End-to-end ViT cats-vs-dogs ML project with training/evaluation scripts, FastAPI inference, CLI prediction, and optional Streamlit UI.

5 pytest tests
GitHub Actions CI
REPORT.md
PyTorchTransformersHugging FaceFastAPIStreamlit

Additional project pages

Supporting work shown with clear scope, repo status, and review notes.

Skills

The stack behind the projects

Grouped around the work I want the portfolio to demonstrate: applied AI implementation, backend systems, data pipelines, ML demos, and delivery quality.

Capability map

Applied AI

structured outputs
agent role boundaries
provider abstraction
optional OpenAI layers
prompt contracts
human review workflows
deterministic fallbacks
selected_capability = "applied ai"

output: structured outputs + agent role boundaries + provider abstraction + optional OpenAI layers

CV package

Education, credentials, and CV links

The CV section is kept as a concise proof block: formal AI education, relevant bootcamp training, downloadable CVs, and credentials that support the portfolio projects.

English and Greek CVs
Bootcamp certificate
BSc + MSc AI background

Junior Software Commissioning Engineer Bootcamp

Code.Hub / LAS Solutions

Bash, Oracle SQL/PLSQL, C++, Docker and warehouse-process simulation work.

Certificate

MSc Artificial Intelligence & Business Strategy

Aston University

AI, machine learning, strategy, entrepreneurship, and decision-making.

BSc Artificial Intelligence & Computer Science

University of Birmingham

AI, algorithms, Java, C/C++, computer vision, security, networks, and CS foundations.

English C2

University of Michigan

Professional English working proficiency.

Contact

Open to AI implementation and backend/data roles.

Best fit: practical AI workflows, automation, reporting, data pipelines, and software systems where the output needs to be understandable and testable.