↑
About Certifications Experience AWS Showcase Music Pipeline Resume Connect Built with Claude ↗
// ENGINEER · DEVELOPER · BUILDER

Cory
Trast

LinkedIn Profile View AWS Showcase
// About

Building real things
on real infrastructure.

I'm a software developer and AWS practitioner based in Leo, Indiana. I learn by building — every project I take on ships real infrastructure, real code, and real solutions.

I'm the engineer and developer behind an AWS-native platform currently in pre-launch stealth. I own the full technical stack — from architecture and data modeling to API design, payment integration, and deployment.

My approach is methodical: plan the work, then work the plan. Whether it's designing a DynamoDB schema, authoring a policy document, or standing up a streaming data pipeline — I document as I build.

I bring deep QE experience to emerging technology — including validation frameworks for GenAI and AI/ML models, API pipeline testing, and data quality engineering across the full AWS stack.

100%
AWS-native stack across all projects
WIP
AWS-native platform — pre-launch stealth
Leo,
Indiana — building from the Midwest
DEA
AWS Data Engineer — Associate (In Progress)
// Certifications

AWS Credentials

AWS Certified Developer Associate badge
AWS Certified Developer — Associate
Amazon Web Services · DVA
Active
Active: 2024-08-23
Expires: 2027-08-23
✓ Verify Credential
AWS Certified Solutions Architect Associate badge
AWS Certified Solutions Architect — Associate
Amazon Web Services · SAA
Active
Active: 2023-07-31
Expires: 2026-07-31
✓ Verify Credential
AWS Certified Cloud Practitioner badge
AWS Certified Cloud Practitioner
Amazon Web Services · CLF
Active
Active: 2022-02-01
Expires: 2027-08-23
✓ Verify Credential
AWS Certified Data Engineer Associate badge
AWS Certified Data Engineer — Associate
Amazon Web Services · DEA
In Progress
Hands-on pipeline build underway
as active exam prep
// Experience

What I've Built

2024 — Present
Stealth Project
Engineer & Developer — Pre-Launch Platform
Designed and deployed a full AWS-native platform from scratch: Lambda@Edge access control with magic-link authentication, multi-role RBAC system, DynamoDB data modeling, API Gateway layer, Stripe payment integration with authorize-at-order / capture-at-delivery flow, SES transactional email, KMS encryption, and a complete observability stack on CloudFront.
CloudFrontLambda@EdgeDynamoDB API GatewaySESKMS StripeAWS Bedrock
2025
Personal Project
AWS Data Pipeline — Showcase Build
Designed and deployed a full end-to-end data engineering pipeline as hands-on preparation for the AWS DEA-C01 certification. The pipeline ingests streaming data via Kinesis, processes it through Lambda, stores it in S3 using a medallion architecture (Raw → Curated → Validated → Anomaly), and exposes results via an API Gateway + Cognito-secured endpoint with a CloudFront-hosted dashboard.
KinesisLambdaS3 Glue ETLDynamoDBCognito API GatewayCloudFront
// AWS Showcase

Data Pipeline Showcase

A fully deployed, end-to-end AWS data engineering pipeline built to demonstrate real-world proficiency across the AWS data stack. Every service is production-configured — this isn't a tutorial clone, it's an architecture built from first principles.

🔐 Live Pipeline Access — By Request

The dashboard and API are live but gated behind AWS Cognito authentication. To be added as a user and receive a walkthrough of the running pipeline, send a connection request via LinkedIn. Once approved you'll have direct access to the live pipeline dashboard.

Request Access on LinkedIn →
// DATA LINEAGE — ACTUAL PIPELINE
⚡
pipeline-trigger
Lambda · Orchestrator
Reset DynamoDB
counts → 0 (TTL 4hr)
invoke async ×2
→
🏭
pipeline-producer
Lambda · 700 records
{ color, value,
timestamp }
10 rec/sec · ~70s
→
📡
pipeline-stream
Kinesis · PartitionKey=color
RED · WHITE · BLUE
streamed records
base64 encoded
→
🔀
pipeline-router
Lambda · Consumer
Decode · batch
by color → S3 raw/
+ DynamoDB ADD
→
🪣
S3 — raw/
JSON · raw/{color}/{ts}.json
↕
🗄️
DynamoDB
pipeline-counts · atomic ADD
→
⏱️
glue-trigger
Lambda · waits 90s
↓
🔧
Glue Workflow
merge-job + anomaly-detection
↓
🪣
S3 — curated/
Parquet · partitioned by color_source
→
📊
get-counts
Lambda · scans DynamoDB
↕
🩺
get-status
Lambda · CloudWatch + Glue
→
🔐
API Gateway
+ Cognito Auth
{ RED, WHITE, BLUE
counts + total }
pipeline status
→
🌐
Dashboard
CloudFront · Live UI
Live color counts
pipeline status
Cognito-gated

How it works

01 Trigger & Reset

pipeline-trigger Lambda orchestrates the run — it resets all three color counts in DynamoDB to zero (with a 4-hour TTL), then asynchronously invokes both pipeline-producer and pipeline-glue-trigger in parallel.

02 Produce & Stream

pipeline-producer generates 700 synthetic records — each with a random color (RED/WHITE/BLUE), a random value suffix, and a timestamp — pushing them to Kinesis at 10 records/sec over ~70 seconds, partitioned by color.

03 Route & Store

pipeline-router consumes the Kinesis stream, decodes each base64 record, batches by color, and writes timestamped JSON files to S3 raw/{color}/. Simultaneously, it atomically increments per-color counts in DynamoDB using ADD expressions.

04 Glue ETL → Parquet

pipeline-glue-trigger waits 90 seconds for the producer to finish, then starts the Glue workflow — running a merge job and anomaly-detection job that read raw JSON, union all color DataFrames, and write columnar Parquet to S3 curated/.

05 Query & Monitor

pipeline-get-counts scans DynamoDB and returns { RED, WHITE, BLUE, total }. pipeline-get-status checks CloudWatch log streams and Glue job run history to report live status of each pipeline stage — producer, router, merge, and anomaly jobs.

06 Dashboard

Both query Lambdas are exposed via a Cognito-authenticated API Gateway endpoint. The CloudFront-hosted dashboard polls both APIs to display live color counts and per-stage pipeline status. Access is by request — reach out via LinkedIn.

Let's connect.

I'm always open to conversations about AWS architecture, data engineering, marketplace platforms, or what I'm building with MUF.

LinkedIn — /in/corytrast View Resume View the Pipeline