Dylan Elder

I build enterprise products and AI workflows that keep people in the decision.

At Swish Analytics, I own NBA and college-football pricing API roadmaps used by 25+ client brands, after starting with tools for in-house traders. Through Agency Edge, I personally build AI research and CRM workflows on the OpenAI API for non-technical teams, with staff confirming every change before it reaches Salesforce.

View work View résumé

Work

Swish Analytics

Product Manager · Dec 2022–Present

Built for expert users first, then for 25+ client brands

I started with Swish’s in-house traders, turning their requests into changes to our trading platform. That work helped cut staffing from 2 traders per game to 1, so Swish could offer clients twice as many in-play games. Now I own the NBA and college-football pricing API roadmaps, where I look for the shared capability behind any one client’s request.

2× in-play games offered

Trader staffing from 2 per game to 1

0 → 7 live customers

September 2025 launch to September 2026

Swish CFB In-Play Player Props API documentation showing game and statistic filters and the modifiedAtMicroGreater update parameter.
View Swish work

Agency Edge

Founder · Part-time alongside Swish

AI proposes. People decide.

I turn manual research, categorization and data entry into reusable AI workflows for non-technical outreach teams. I build them on the OpenAI API with structured outputs and function calling. Teams set the confidence level for auto-approval, review everything below it, and confirm before anything is written to Salesforce.

88% / 12% AI-proposed changes

Accepted as written / edited or rejected in review

2.6% → 0.6% bounce rate

Photography talent agency campaigns, 2024 vs. 2025

160 → 5,078 records

Marin Academy, from pilot to paid engagement

Agency Edge review queue: an AI-proposed Role Type change at 67% confidence, showing the new value, the current value, and Approve and Reject controls.
View Agency Edge work

How I build

Earlier venture

Dylan Elder wearing a blue collared shirt outdoors

About me

I work best close to the people doing the work: learning how they actually work, making clear product choices, and staying involved through launch and adoption.

Since 2023, ChatGPT and now Codex have let me go from writing specs to building and shipping software myself, for real customer teams through Agency Edge.

I left Georgetown to build Monster Roster, then returned to finish my psychology degree (2019–2022).

Let’s talk.

I’m looking for product roles building AI for knowledge workers, working closely with customers, engineers and research teams to turn complex workflows into products people trust.

LinkedIn