The platform team you don't have yet.
Cloud infrastructure, security and compliance, data engineering and AI, from one senior engineer. Seven years building production systems on AWS, most recently a multi-year migration of legacy SQL Server workloads into Spark at a Fortune 100 telecom. You work with me directly: no bench, no account manager, no handoff to someone junior.
Four kinds of problems I get called for.
Cloud & platform infrastructure
AWS, GCP and Azure set up to stay maintainable: Terraform, CI/CD, containers, monitoring and cost control. From a first production environment to the on-prem migration you keep putting off.
Security & compliance
Least-privilege IAM, secrets, network and audit logging, and code and dependency review. Then the technical controls HIPAA, CMMC and FedRAMP require, mapped to your actual configuration, with the evidence an assessor will ask for.
Data engineering
Pipelines, warehouses and lakehouses that hold up unattended. SQL Server to PySpark, on-prem to cloud, Airflow, dbt and Iceberg, migrated so the business keeps reporting while the platform moves underneath it.
AI implementation
LLM workflows on your own data, such as document extraction, internal assistants and MCP servers, built on Claude, Codex or Amazon Bedrock with access controls from the start. Plus AI coding assistants rolled out to a dev team with guardrails and usage limits.
What the work has been worth.
From engagements at a Fortune 100 telecom, a global agribusiness and a loyalty-marketing platform. Client names withheld; the numbers are theirs.
Annual licensing cost redistributed through a Cloudera-to-AWS migration, freeing budget for platform work.
Deployment cadence, moved in two steps through automation and template-driven pipelines. Deployment cycles ran 25% faster.
A critical legacy processing workflow that took multiple days, rebuilt as Spark on EMR. Same output, overnight instead of over a weekend.
Licensing saved over 18 months by moving off enterprise ETL tooling onto the open-source equivalent, without a reporting gap.
New recurring revenue from fuzzy matching and similarity scoring applied to records the business had already collected.
Manual analysis eliminated by sentiment analysis workflows, returning a week of analyst time every week.
The first two weeks.
Projects and retainers start the same way, whether they run a month or two quarters. A fixed-price starter runs the same steps inside its own hours. You know what you're getting before anything is committed.
30-minute call
You describe the problem. I tell you whether I'm the right person for it, and if I'm not, what kind of help you actually need. No deck, no discovery fee.
Read the system
Repos, cloud accounts, pipelines, and the three dashboards people actually trust. I write down what's fragile, what's exposed, what's expensive, and what's fine despite looking alarming.
Plan, then ship something small
A written sequence with owners and estimates, ordered by risk rather than by what's easiest to demo. Then one real change in production, so the plan is proven and not just plausible.
Work in your tools, leave documentation
Your repo, your board, your standups. Everything I build is documented and handed over as it lands, so your team can run it without me — and so ending the engagement is never a cliff.
Start small, priced up front.
Starter
10–20 hrs · fixed priceOne defined problem, a fixed price agreed on the first call, and a written next step at the end. The usual ones are listed below.
Best when you want to see the work before committing to more.
Retainer
From 10 hrs / monthA senior engineer on your platform every month: the backlog, the security findings, the next migration. For teams carrying more platform work than headcount, or holding a role open while they hire.
Best when the work is continuous and priorities move week to week.
Project
Fixed scopeA migration, a platform build, a compliance remediation, or an architecture review with a sequenced plan. Defined deliverable, defined finish line, fixed or milestone-based fee.
Best when you know what needs to exist and need someone to build it.
For IT firms
Under your brandOverflow and specialist work for MSPs, dev shops and data consultancies: the AWS project, a client's compliance evidence request, the AI question nobody on the team has time for. Delivered under your name, at a partner rate.
Best when a client needs a skill your team doesn't have this month.
Starter packages
Fixed price once the scope is agreed on the call. Each one ends with a written next step, which you're free to take elsewhere.
AWS security & cost review
Ranked findings across IAM, network, logging and spend, with the top three fixes applied.
~12 hrsfrom $2,400Compliance readiness check
HIPAA, CMMC Level 2 or FedRAMP: the technical controls you're missing, mapped to your configuration, with the fixes in order.
16–20 hrsfrom $3,200AI pilot on one workflow
A working prototype on your own data, such as document extraction, triage or an internal assistant, plus a written go / no-go.
~20 hrsfrom $3,900AI coding-assistant rollout
Claude Code or Codex set up for your team: settings, guardrails, usage limits and a short team guide.
~12 hrsfrom $2,400Pipeline or reporting fix
The broken feed fixed, with tests and alerting added so it stays fixed.
10–15 hrsfrom $1,900Who this works for.
What I work in.
Tools I use week to week, not a list of everything I've touched. The modeling work sits next to the engineering — most of the results above came from doing both.
Eight years in the U.S. Army Reserve as a CBRN specialist. Borealis Analytics Corporation since 2025.
Tell me what you're building.
A 30-minute call, one business day to reply, and a straight answer on whether I can help. If I can't, I'll say so on the call.