Tej Shah

Work

I’m a soft­ware engi­neer at Amazon Web Ser­vices in New York, work­ing on AWS Billing and Cost Man­age­ment. I build and oper­ate the dis­trib­uted sys­tems and APIs behind Cost Explorer, the Billing Con­sole, and the Sav­ings Plans and Reserved Instance reports: ana­lyt­ics plat­forms that process petabytes per day and help mil­lions of cus­tomers manage hun­dreds of bil­lions of dol­lars in cloud spend world­wide.

I led engi­neer­ing for Bedrock cost attri­bu­tion, coor­di­nat­ing four teams to extend five Cost Explorer APIs and ship the public SDK, so cus­tomers can group, filter, and budget gen­er­a­tive-AI spend by model, provider, and infer­ence type. I also helped ship Data­base Sav­ings Plans, announced in the re:Invent 2025 key­note, and upgraded the 50 pro­duc­tion search clus­ters behind Cost Explorer with min­i­mal down­time, scal­ing multi-payer billing views to 4x the traf­fic while cut­ting p99 latency 28%, improv­ing data fresh­ness 70%, and adding three years of runway.

I lead gen­er­a­tive-AI adop­tion for oper­a­tions across my orga­ni­za­tion, built as three stacked logis­tic curves that stay on an expo­nen­tial: as one sat­u­rates, the next takes over. The first turned resolved inci­dents into search­able knowl­edge for 75 engi­neers on 10 teams, cut­ting weekly high-sever­ity inci­dents from 50⁠–⁠80 to about 25 and MTTR by 30%. The second, now built out, gives multi-agent sys­tems shared con­text and reus­able skills. The third and last step is RL envi­ron­ments and post-train­ing. I also led the team pro­gram that cut our ticket queue from 500 tick­ets to 50, which received my orga­ni­za­tion’s Q2 2026 AWS Oper­a­tional Excel­lence award.

Before AWS, I did research at Rut­gers com­par­ing lan­guage models with cog­ni­tive models of human memory, and I was Kleiner Perkins’ first under­grad­u­ate Prod­uct Fellow, work­ing as a prod­uct man­ager at Interos. I stud­ied com­puter sci­ence and busi­ness ana­lyt­ics at Rut­gers as a Pres­i­den­tial Scholar and grad­u­ated summa cum laude and Phi Beta Kappa. I’m also a co-author of Human­ity’s Last Exam in Nature.

Experience

  • Amazon Web Services

    2024 to present
    Seattle and New York

    Software Development Engineer, Billing and Cost Management

  • Rutgers University

    2021 to 2024
    New Brunswick

    Research Engineer, Assistant and Intern

  • Kleiner Perkins Fellows

    2022
    San Francisco

    Product Fellow, Product Manager at Interos