- Lead the Transactional Storage org (9 teams, 66 engineers), owning Datadog's application data storage platforms: Postgres, UnifiedKV (the in-house key-value database), Search, Data Replication, and Caching.
- Cut seven figures a year in infrastructure cost in a single quarter: migrated Postgres shards across 6 data centers to shared compute in under 4 weeks, deleted unused staging clusters, and right-sized production fleets.
- Set direction for application data storage: retired two legacy database platforms, consolidating onto the internal KV database and Postgres PaaS.
- Reset a stalled rewrite of the in-house KV storage engine and shipped it to GA: 2× load absorbed in <10 minutes, 30% lower cost, and better performance at scale.
- Matured Frames, the internal platform that keeps product data flowing through downstream failures: 20+ new use cases, higher intake availability. Folded it into Caching to remove product overlap.
Engineering Leader · New York
I build things people use, and the teams that ship them.
These days I lead Transactional Storage at Datadog. I like hard problems, small teams, and shipping.
About
I'm a mechanical engineer who fell for software.
Out of Georgia Tech I wanted to build physical things. Things people would use. I started at Quirky building prototypes, went to Wink to take products apart and learn how they worked, and somewhere in there I learned to actually code. Then to run the infrastructure underneath. Then to lead the teams that build it.
The title kept changing. What I do didn't. I build things, I break things, and I work out how the pieces fit. These days that means leading engineering at Datadog: building the teams, launching the products, and owning what they earn. It's the same job I had on day one, just bigger.
I'm happiest where software meets the physical world: sensors, hardware, data about something actually happening out there. And I care a lot about being the kind of manager people do their best work for.
How I operate
I'm a technologist: I point technology at people problems and engineering problems. I don't like being bored.
Build with customers.
Products earn their keep. Data Warehouse Monitoring went from two engineers to a paying customer in four weeks. Data Streams went from pre-revenue to multi-million-dollar ARR in a year because we built it next to the people using it.
Stay close to the computers.
I read the code, sit in the incident channel, and ship my own products on nights and weekends. Managers who drift from the work make worse calls about it.
There's always a way.
A stalled storage engine rewrite became a GA launch. A Postgres fleet moved across six data centers in under four weeks. Most impossible problems are focus problems.
Execute short, aim long.
Stay focused on the quarter with a clear view of the vision. Ship, learn, adjust, and repeat until it sticks.
First 90 days anywhere: meet the customers, read the code, find the money, ship something.
Experience
- Led engineering for the APM Data Observability suite (14 engineers).
- Incubated and grew two new products (Data Streams & Data Jobs Monitoring) for the data-engineer persona, winning new accounts.
- Scaled Data Streams Monitoring from pre-revenue to multi-million-dollar ARR in 12 months.
- Took Data Jobs Monitoring from private beta to GA: double-digit month-over-month revenue growth in its first 6 months.
- Prototyped Data Warehouse Monitoring with 2 engineers, securing the first customer within 4 weeks.
- Doubled the team in 12 months across 180+ interviews while keeping a startup-like, customer-obsessed operating model.
Built by Experiment nights & weekends
- A lab for staying sharp on 0-to-1: three AI-first SaaS products shipped solo on Lovable and Supabase.
- txt.wedding: SMS announcements for couples, live with active users.
- Green Thumb: weather-smart yard and plant care reminders for homeowners.
- Caring Loop: care coordination for family caregivers.
Numina
- Owned engineering, product, and R&D for edge-AI computer-vision sensors, the data platform, public API, and analytics dashboard, plus custom hardware.
- Drove an accuracy initiative (model retraining + algorithm work) that improved data accuracy 20–50%.
- Revamped code delivery and testing, cutting shipping time from months to under a week.
- Designed a new data lake, pipeline, and landing zone to absorb 3× ingestion growth.
- Cut time-to-hire from 6 months to 4–8 weeks and built a team spanning hardware, firmware, ML/CV, and full-stack.
- Led five infrastructure teams across Compute, Traffic, Storage, Data Center, and Security.
- Architected a performance rewrite of the core edge load balancer: 5× faster secure connections on every request.
- Migrated 3,000 VMs to bare-metal Kubernetes: deploys 1.5h → 5min, $100k/yr saved, 2× faster incident recovery.
- Drove a 3× gain in data-center efficiency (60%+ utilization) and owned $5M of capital procurement.
- Integrated pre-IPO SOX controls into infrastructure; primary Incident Commander and Code Camp instructor.
Earlier
- Kinoko, technical co-founder (side project): peer-to-peer video chat for the aging population. WebRTC, Go, TypeScript, Postgres, and a Raspberry Pi TV kiosk.
- Wink, site reliability engineer: migrated 18 applications from Heroku to AWS Kubernetes in three months and championed SRE culture across engineering.
- Quirky, systems engineer, then BlackOps design engineer: prototyping specialist on the CEO's four-person design team. Built the prototypes for Quirky Aros and Wink Relay, and led Asia vendors through final R&D for new smart-home products.
Full history with dates and detail lives in the résumé (PDF).
Contact
Let's talk.
Talk to me about senior engineering leadership, advisory work, and board seats.