High-Stakes Reporting & Compliance
Organizations in healthcare, finance, or aerospace that require absolute traceability, rigorous validation, and audit-ready data flows.
Bridging Silicon, Cloud, and Applied AI Research for Regulated Industries
Reliable Data Systems. Rigorous Engineering. Applied AI Research.
Decades of hands-on engineering in regulated, high-consequence environments — clinical data and imaging at pharmaceutical scale, cloud-native platforms, and audit-ready reporting. Alongside that consulting practice, I independently research where deterministic engineering and probabilistic AI reasoning should meet, publishing governed, reproducible findings rather than treating either as settled. Open to consulting, contract, and technical-leadership engagements across both.
Principal lead & embedded builder
Audit-ready compliance (GXP/Healthcare)
IoT & Embedded Systems (C++/ESP32)
Agentic AI Workflows (OpenClaw)
GCP & AWS Cloud Architecture
The "Best Fit" for My Expertise
I specialize in environments where "close enough" isn't an option. My work is most impactful for:
Scaling Startups & Modernization
Teams outgrowing spreadsheet-heavy reporting who need a practical, hands-on lead to build a governed analytics foundation.
Edge-to-Cloud (IoT) Initiatives
Companies bridging the gap between hardware (ESP32/Embedded) and the cloud, requiring unified engineering across the entire data lifecycle.
AI-Ready Infrastructure
Teams looking to integrate agentic AI not as a gimmick, but as a robust operational layer for automated pipeline management and diagnostics.
What I Help Teams Deliver
Governed Analytics Platforms
Move critical business logic from Excel, Google Sheets, and one-off SQL into governed warehouse models, orchestration, and documented operating practices.
Agentic Cloud Operations
Leverage agentic augmentation to automate dbt and BigQuery pipeline configuration, reducing manual overhead and improving management at scale.
Cloud Architectures That Hold Up in Production
Design event-driven, cloud-native platforms on GCP or AWS using practical patterns that balance delivery speed, maintainability, observability, and long-term cost control.
Audit-Ready Regulated Data Workflows
Build auditable ingestion, transformation, and delivery systems for healthcare and other high-stakes environments where traceability and validation cannot be treated as afterthoughts.
Why Companies Hire Me
Trusted Numbers
Reliable pipelines, governed transformations, and clearer ownership help teams make revenue and operating decisions with less debate.
Faster Delivery
I work hands-on across architecture, implementation, debugging, and operations, which shortens the path from diagnosis to production improvement.
Cost Awareness
Platform choices are tied to business value, cloud spend visibility, and maintainability, not abstract architectural fashion.
Representative Engagement Outcomes
- Agentic dbt/BigQuery management: Implemented agent-augmented workflows for automated dbt model generation and BigQuery partition management, significantly reducing configuration drift.
- AI-assisted validation: Explored LLM-assisted anomaly detection and metadata-aware validation to accelerate triage and improve reliability in regulated data pipelines.
- Agentic systems exploration: Built local agent-based environments using OpenClaw to explore persistent context, tool invocation, and multi-step reasoning for engineering automation.
- Reliability remediation: Built an early-warning framework with Airflow, dbt tests, BigQuery metrics, and Slack alerts so pipeline failures surfaced before they became executive surprises.
- Cloud ingestion hardening: Designed Gen2 Cloud Functions based ingestion services with validation, schema control, and robust handling for bursty workloads.
Experience Snapshot
AI & Agentic Systems Exploration (2025-Present)
Integrating LLMs into engineering workflows using OpenClaw for structured output generation, pipeline diagnostics, and automated validation frameworks.
Podimetrics (2024-2025)
Principal quality and platform engineering for GCP. Explored LLM-assisted validation and anomaly detection to accelerate triage in regulated healthcare pipelines.
SURL Founder (2023-2024)
Designed and built secure AWS transfer infrastructure with Kafka, Python services, PostgreSQL, Terraform, observability, and explicit cloud cost accountability.
Pfizer (2009-2023)
Senior technical leadership across clinical data processing, DICOM archive search, data anonymization analysis, modernization, and validated systems in regulated environments.
Where I Have Built
Core Technologies
Frequently Asked Questions
What engagement models do you support?
I support a range of needs including project-based consulting, fractional leadership (Head of Data/Platform), contract-to-hire engagements, and permanent leadership roles for organizations looking for long-term ownership.
What engagement shape works best?
Most projects start with an assessment or a narrowly scoped remediation effort, then expand into implementation, fractional leadership, or permanent placement if the fit is right.
What kinds of stacks do you usually improve?
Typical stacks include Python, SQL, Airflow, dbt, BigQuery, PostgreSQL, Kafka, Cloud Functions, Cloud Run, Snowflake, AWS infrastructure, and observability tooling.
How I Work
Diagnose Fast
I start by locating the trust gaps, operational risks, and architectural constraints that are actually slowing the business down.
Lead With Practicality
I favor phased improvements, clear ownership, and operating discipline over abstract redesigns that are expensive to adopt.
Build and Handoff
I work hands-on through implementation and stabilization, then leave behind architecture clarity, monitoring, and a team that can keep moving.
Research
Independent Research
My research investigates the boundaries between what a system can determine, what it can infer, and what it can independently verify — currently a proposed direction on deterministic and probabilistic system integration, building on two prior, independently scoped programs.
Proposed — Not Yet Started
Deterministic Boundaries for Probabilistic Systems
When can probabilistic systems provide measurable utility beyond deterministic approaches, and what governance does that require? The first proposed experiment — comparing deterministic, lexical, semantic, and hybrid retrieval over structured DICOM metadata — has not yet been conducted.
Phase I Closed
Evidence Packets
Representing otherwise-equivalent evidence as an Evidence Packet showed no demonstrated correctness advantage. A follow-on comparison of durably captured versus later-reconstructed evidence found a threshold-crossing but statistically thin advantage for durable capture.
Research Milestone Complete
DICOM Trust Boundary Research
Demonstrated that pre-persistence structural parsing and policy can hold across four escalating persistence contexts, on real DICOM objects — a completed engineering milestone, not a production or compliance claim.
Latest from the Blog
September 2026
When Better Evidence Doesn't Produce Better Answers
Evidence Packets Phase I closes: a clean null on representation, and a thin, qualified positive signal on preservation.
Read ArticleSeptember 2026
Before It Lands
A demonstrated pre-persistence DICOM policy boundary, pushed through five escalating boundaries against a real object store.
Read ArticleNeed a Senior Partner for Modernization, Reliability, or Regulated Data Work?
If your team is wrestling with pipeline instability, spreadsheet bottlenecks, metric trust issues, or regulated data complexity, I can help assess the problem, define a practical roadmap, and lead the implementation.
kris.kokomoor@gmail.comBest fit: startups scaling analytics, teams that need a 3- to 6-month technical lead, healthcare data organizations, and companies with high-stakes reporting or compliance requirements.