manish thakrani

Building the next generation of identity verification

Built a mutli tenant, highly available observability platform that allows for service and infrastructure monitoring across Abnormal's commercial and federal deployments. Set up a SRE function in the company to own service reliability, incident response and CI/CD foundational platforms at the company. Currently focussed on delivering enhancement to critical systems powering the core email scoring and remediation products to improve to 4 9's of availability

Led Twilio’s global IAM Authentication Org, delivering secure unified identity infrastructure across Twilio and major acquisitions (Segment, SendGrid, Electric Imp), enabling consistent user experience and faster integration. Launched a centralized Authentication control plane built on Auth0 to allow for unified identity workflows on Twilio, Sengrid and Segment which improved the company's overall cross and upsell momentum and also reduced new composite product development by 40% Launched a centralized RBAC platform, empowering customers with fine-grained access control and reducing support load by ~80%, while accelerating compliance readiness for enterprise audits (SOC 2, ISO 27001) Built GDPR-compliant, low-latency IAM infrastructure in Ireland and Australia, cutting authentication latency by over 50% for international users and enabling regional data residency for regulated markets. Drove IAM roadmap and execution, securing multi-million dollar investment and achieving 99.99% availability and strong security posture (near-zero CVEs) through rigorous engineering practices. Scaled and developed a high-performing team, growing headcount from 8 to 25 across time zones, maintaining strong eNPS, and promoting multiple leaders into management roles.

Scaled a high-performing org from 7 to 100+ engineers across Menlo Park, Houston, and Pune in just two years—establishing distributed agile practices and a strong engineering culture. Pioneered machine-learning platforms for seismic modeling and energy-production forecasting, cutting interpretation cycles and informing multimillion-dollar field-development decisions. This infrastructure reduced new drilling site analysis time by 50% Built the MLOps backbone (data-cleansing pipelines, experiment lineage, push-button inference deployments) that slashed model release times from weeks to hours and ensured full auditability. Architected a poly-cloud Delfi stack on GCP + Azure, delivering high availability, cloud-vendor flexibility, and compliance with global data-sovereignty requirements (https://www.slb.com/videos/delfi-environment) Drove strategic tech integrations—analyzing and onboarding solutions from energy-sector VC portfolio companies to expand Delfi’s capabilities and accelerate time-to-value for customers.

Part of the Video Platform group (which successfully streamed the first ever free global NFL live stream). Lead on the elastic transcoding and video analytics platforms Modernized live video streaming infrastructure, replacing legacy RTMP pipelines with HLS-based delivery, improving playback reliability and device compatibility at scale. Implemented just-in-time packaging for on-demand video into HLS/HDS formats, significantly reducing storage overhead while maintaining flexible, device-specific delivery. Optimized global content delivery via enhancements to distributed edge caching, DNS-based load balancing, and proxy subsystems, improving cache hit ratios and reducing user latency. Built an elastic video transcoding system using OpenStack and FFmpeg, enabling on-demand scalability for video workloads with cost-efficient resource utilization.