Portfolio Deck — 2026Data Engineering

Slide 01 · Introduction

Keith AaronBoccio

Data Engineer Lappeenranta, Finland

I build the pipelines, lakehouses and streaming systems that turn raw events into data people can actually trust.

Keith Aaron Boccio
K. A. BoccioNavigate ↓→
About02

Slide 02 · About

Plumbing
for data.

I'm a data engineer with eight years of experience designing data platforms for logistics, energy and e-commerce teams. My work sits between software engineering and analytics: I take messy operational sources and turn them into reliable, documented, well-modelled datasets.

I care about boring things done well — idempotent jobs, schema contracts, tested transformations, sensible costs — because that's what lets analysts and ML teams move fast without second-guessing their numbers.

8+years in data
40 TBprocessed daily
99.9%pipeline SLA
Toolkit03

Slide 03 · Skills

The
stack.

Pragmatic tooling, chosen for the problem rather than the hype.

A

Languages

PythonSQLScalaBash
B

Processing

Apache SparkKafkaFlinkdbt
C

Orchestration

AirflowDagsterGreat Expectations
D

Storage

SnowflakeBigQueryPostgreSQLDelta LakeIceberg
E

Cloud & Ops

AWSGCPTerraformDockerKubernetesCI/CD
Case 01 / 0304

Slide 04 · Project

Fleet telemetry in real time.

A streaming platform ingesting GPS and sensor events from 6,000 delivery vehicles, enriching them on the fly and serving live ETAs to dispatch dashboards.

Events / sec
120k
End-to-end
< 4 s
Infra cost
−35%
KafkaFlinkIcebergKubernetesGrafana
KAFKA FLINK DASHBOARD ICEBERG VEHICLES
Case 02 / 0305

Slide 05 · Project

Warehouse to lakehouse.

Migrated a 12-year-old on-prem warehouse to a cloud lakehouse with medallion layers, dbt models and data contracts — without a single day of reporting downtime.

Tables moved
1,400
Nightly batch
6h → 40m
Tests added
3,200
SnowflakedbtAirflowTerraformAWS S3
BRONZE · raw SILVER · cleaned GOLD · marts
Case 03 / 0306

Slide 06 · Project

Data quality you can see.

An internal observability layer tracking freshness, volume and schema drift across 900 datasets, with automated lineage-aware alerts routed to the owning team.

Datasets
900
Incidents
−62%
Time to detect
9 min
PythonGreat ExpectationsDagsterOpenLineagePostgreSQL
INCIDENTS / MONTH — AFTER ROLLOUT −62%
Track record07

Slide 07 · Experience

Where I've
shipped.

Eight years across platform teams, from first data hire to senior engineer.

2022 — Now
Senior Data EngineerNordic logistics group

Lead the streaming and lakehouse platform; own data contracts, mentor four engineers and drive cost governance on cloud spend.

2019 — 2022
Data EngineerRenewable energy analytics

Built Spark pipelines for turbine and grid sensor data, introduced dbt and CI testing, and cut batch runtime by 80%.

2017 — 2019
BI & ETL DeveloperE-commerce scale-up

First data hire: designed the warehouse schema, automated reporting ETL and set up the company's first Airflow deployment.

2013 — 2017
B.Sc. Software EngineeringTechnical university

Focus on databases and distributed systems; thesis on incremental view maintenance.

Final slide08

Slide 08 · Contact

Let's talk
data.

Send an email → Call