Dylan Kane Elder

Technical product manager building enterprise APIs and AI workflows.

San Francisco, CA · dylankelder@gmail.com

Experience

Product Manager · Swish Analytics

December 2022–present

NBA and college-football roadmap owner since early 2025. I lead real-time enterprise API products across pricing, settlement, integration, and operational controls.

  • Owned 30+ product releases for real-time enterprise APIs, setting roadmap priorities, defining product requirements, and leading cross-functional delivery from customer needs through production launch.
  • Set CFB In-Play as the top roadmap priority; led player, team/match, and alternate-line APIs through staged release, meeting customer-required milestones and enabling multiple major operators to go live.
  • Lead client-facing resolution of NBA Fast Props API integration issues using Datadog request analysis; coordinate staged engineering fixes and confirm recovery directly with clients.
  • Expanded shared pricing and settlement APIs with 26 CFB team/match markets without structural changes; defined alternate-line contracts preserving identifiers and specifying projections, probabilities, suspensions, and incremental updates.
  • Led staged rollout of NBA selection-level suspension controls, allowing clients to keep eligible selections active independently while preserving legacy market-level suspension behavior.
Swish portfolio overview CFB In-Play case study

Independent Founder & Builder · Agency Edge

August 2023–present

I also built Agency Edge in my spare time. Three paying organizations have used the AI product: a photography agency and two schools.

  • Led technical discovery and personally implemented the React/TypeScript frontend, Python/Flask services, relational data, integrations, and asynchronous jobs. Collaborators contributed commercial work, quality review, and delivery coordination.
  • Adapted the shared application for two paying schools. One delivery reported 70K+ updated field values. The other delivered 156 populated email records across 498 rows with source identifiers preserved, followed by payment and a customer-posted referral.
  • Expanded existing field confidence into contextual identity review and hybrid AI-and-rules scoring after customers found wrong-person matches, then reused the approach across customer workflows.
  • Evolved the application to separate accepted enrichment from eligibility to update a CRM through typed changes, destination mappings, overwrite policies, and stale-verification invalidation.
Agency Edge case study

And1Analytics → Monster Roster

Co-Founder & President · February 2015–June 2016
Founder & CEO · July 2016–December 2019

I led a consumer analytics venture from paid recommendations into subscription web and mobile products. It acquired 1,000+ paying customers and was the first company selected for the Philadelphia 76ers Innovation Lab.

After identical lineups split customer winnings, I led a multidisciplinary team to build a guided optimizer combining algorithm-backed locks, recommendations and user choice.

Monster Roster case study

Education

Georgetown University · Bachelor of Arts, Psychology · 2022

Contact

dylankelder@gmail.com · LinkedIn · GitHub