Background
I'm Ken Flynn — a data engineer with a Ph.D. in microbiology and over ten years building data infrastructure for high-growth software companies. I founded Flynn Data Services LLC in 2026 to bring senior-level data engineering to B2B SaaS companies that don't need (or can't justify) a full-time hire.
My path started in a research lab, where I built computational pipelines to extract signal from genomic sequencing data. That same instinct — reduce dimensionality, find the signal, build something that runs reliably — carried straight into data engineering. Different domain. Same discipline.
I specialize in Snowflake-based data environments, CRM/RevOps pipeline architecture, and cost optimization across cloud infrastructure and SaaS tooling. My clients are typically B2B SaaS companies between $5M–$50M ARR with complex MarTech stacks and limited in-house data resources.
Experience
Founder & Principal Data Engineer — Flynn Data Services LLC
- Founded boutique data consultancy serving B2B SaaS companies with complex MarTech stacks and limited in-house data engineering resources
- Deliver Data Stack Audits identifying cost optimization opportunities across cloud infrastructure, SaaS tooling, and data pipelines for Snowflake-based environments
- Provide Fractional Data Engineering and CRM/RevOps pipeline architecture — on-demand senior expertise for companies without full-time data teams
- Developing BI-in-a-Box: enterprise-grade analytics infrastructure using modern open-source tooling (dlt + dbt + DuckDB + Evidence) at a fraction of traditional costs
Lead Data Science Engineer — Cybrary, Inc.
96% CRM latency reduction 92% Algolia cost reduction 52k invalid contacts scrubbed
- Engineered a custom Reverse ETL engine (Airflow + Cloud Run) syncing Snowflake to HubSpot every 15 minutes — reducing CRM data latency by 96% vs. the prior 24-hour vendor sync
- Engineered backend app-engagement data pipelines for Cybrary's Salesforce-to-HubSpot CRM migration, rebuilding data synchronization to eliminate middleware dependency and support a 66% reduction in annual CRM licensing costs; deprecated Hightouch integration by building a custom solution
- Right-sized Algolia contract from Enterprise to GROW tier after auditing actual search volume, reducing annual search costs by 92% with no change in performance
- Architected a high-velocity HubSpot Ingress Engine (Python/Cloud Run) processing 32k+ monthly signups with idempotent upsert logic and real-time email validation (52k+ invalid contacts scrubbed at 9.4%)
- Implemented GDPR/CCPA-compliant consent management via Google Tag Manager and iubenda (43 tags configured); restored attribution for 30–40% of enterprise SSO signups by engineering custom UTM/Gclid header extraction in the PHP Auth API during SSO handshakes
- Introduced repository-wide pytest standards with dslib-powered GCP service mocks, shifting from manual production testing to automated CI gates catching 3–5 critical regressions per month
Senior Data Engineer — Living Security, Inc.
- Designed and architected the Events API enabling customers to send event data directly to the Unify platform, bypassing third-party access requirements
- Led development of Unify's Behavior Score model, summarizing user risk in a human-readable and customizable format
Senior Data Engineer — Cybrary, Inc.
- Architected Cybrary's first centralized reporting and data analytics system aggregating data for millions of users across multiple sources
- Oversaw migration from AWS to GCP and maturation of the system into a reliable, self-serve platform as user count grew
- Built an ETL pipeline routing thousands of leads per day for marketing and sales campaigns in near real-time
- Led closed-captioning initiative for 74,000 audio minutes of video content; designed a quality-scoring algorithm that increased average caption quality score by 29.7%
Field Data Scientist — Forcepoint / RedOwl Analytics
- Worked with regulatory surveillance and information security customers to design analytical strategies for insider risk use cases
- Developed ETL pipelines enabling both the data science and sales teams to pursue new analytical opportunities using novel data sources
- RedOwl Analytics was acquired by Forcepoint in 2017; joined the Professional Services group to help sell, implement, and support UEBA deployments
Postdoctoral / Graduate Researcher — University of New Hampshire
- Built genomic sequencing data pipelines (Trimmomatic, bowtie2, breseq) to help researchers extract actionable insight from microbial sequencing data
- Developed an algorithm to reduce the dimensionality of metagenomic datasets by up to 99.89%, directing limited compute to the most functionally important mechanisms
Education
Ph.D. in Microbiology — University of New Hampshire
B.A. in Biology — Colby College
Stack
Languages I use day-to-day:
Python SQL R Perl
Infrastructure & tooling:
Snowflake Airflow Docker Terraform dbt DuckDB Looker Fivetran Snowplow Hightouch Apache NiFi UNIX
GCP:
Cloud Run Cloud Functions Pub/Sub GCS Secret Manager Cloud Build
AWS:
Lambda S3 Step Functions SNS SQS API Gateway Secrets Manager IAM Transcribe RDS
CRM/MarTech:
Salesforce HubSpot Pardot Marketo
Work Together
Available for fractional retainer engagements and project work. Typically 2–3 new clients per quarter.