Prem Moola
CTO | Head of Data & AI Platforms | AI Infrastructure | Enterprise Architecture | Ex-Goldman Sachs | BNY Mellon
[email protected] | linkedin.com/in/premmoola | www.premmoola.com
Professional Summary
Technology executive and hands-on platform architect with 20+ years of experience building enterprise AI platforms, data engineering systems, cloud-native SaaS products, and global engineering organizations across financial services, SaaS, media, and founder-led environments. Strong fit for CTO, Head of Data and AI, Enterprise AI Platform, VP/SVP Engineering, AI Transformation, and AI Solutions Architecture roles. Known for modernizing complex platforms, scaling distributed teams, and converting AI/data strategy into production capability. Outcomes include 80% infrastructure cost reduction, 75% operational cost reduction, 40% custom-reporting cycle-time reduction, 70% faster time-to-insight, 5x deployment-velocity improvement, and 99.99% uptime across production SaaS/media workflows.
Core Competencies
Enterprise AI platform architecture; Head of Data and AI strategy; Agentic AI systems; LLM infrastructure; RAG; governed AI adoption; data engineering platforms; cloud-native architecture; platform modernization; distributed systems; Spark; Hadoop; HDFS; Sybase IQ; MemSQL; Elasticsearch; MongoDB; DB2; Kafka; regulatory data platforms; trading data platforms; risk systems; entitlement systems; auditability; data lineage; engineering leadership; global teams; vendor strategy; operating-model transformation.
Professional Experience
- Building a self-hosted enterprise AI platform for private inference, retrieval, governed tools, identity-aware access, and operator workflows.
- Set the core architecture invariant that customer documents, prompts, responses, audit records, identity data, and backups remain inside the customer's infrastructure perimeter, with every outbound channel explicitly operator-enabled.
- Designed srasta-api as the single external enforcement gateway for TLS, OIDC/JWT validation, RBAC, per-role model access, license posture, rate limiting, signed principal headers, and internal service proxying, reducing duplicated authorization logic across services.
- Architected the AI execution plane around OpenAI-compatible RAG and inference APIs, hybrid dense and BM25 Milvus retrieval, tenant-scoped collection filtering, LiteLLM model routing, vLLM GPU inference, Ollama and TEI fallback paths, and MCP-style governed tool execution.
- Productized the Srasta Membrane as the governed memory and state boundary inside AI execution, covering structured memory, branch and commit lifecycle, candidate commits, drift evaluation, policy and approval decisions, model-aware context rehydration, lineage, and auditable rollback.
- Designed a performance-evaluation loop for the enterprise intelligence layer, using persona eval sets, routing accuracy, latency, quality, throughput, cost, hardware fit, and operator constraints to tune model selection, retrieval behavior, prompts, governance policies, and future fine-tuning priorities.
- Framed the ROI model for enterprise AI infrastructure by consolidating fragmented model serving, retrieval, memory, tooling, identity, audit, deployment, and recovery work into one governed platform that can reduce build time, vendor/security-review sprawl, operational risk, and uncontrolled AI spend.
- Established deployment paths for single-node Docker Compose, guided multi-host installs, and Kubernetes/Helm operations across EKS, GKE, AKS, and k3s with GPU placement, external inference, ingress, storage, migration, and rollback options.
- Defined enterprise hardening patterns including hash-chained audit logs, SIEM/export posture, digest-pinned images, SBOM, cosign, Grype release checks, offline license validation, and policy profiles for regulated deployments.
- Building Diggt as a digital product studio and shared platform layer for mobile-first SaaS products across property operations, workforce management, habit systems, subscriptions, analytics, CRM, and cloud cost operations.
- Established Studio Platform as Diggt's shared product foundation across three live products, reusing authentication, notifications, analytics, CRM, deployments, and partial payment capabilities to reduce incremental product build effort by roughly 60%.
- Built the shared AWS/serverless platform using Flutter clients, DynamoDB, S3, EventBridge, API Gateway, Lambda, CDK, Cloudflare Workers/Pages, SES messaging, product tracking, and cost reporting while keeping all products at roughly $20/month in cloud spend.
- Built SureLease, a live cross-market rental property management SaaS for US and India markets, supporting owner and tenant workflows across properties, units, lease/rent operations, onboarding, OTP authentication, subscriptions, notifications, and mobile/web delivery. Website: surelease.diggt.cloud.
- Preparing SureLease beta onboarding for 11 property owners, including Indian and Caribbean property owners domiciled in the US, with focus on digital identity, KYC onboarding, and cross-border rent/payment operations.
- Architecting SureLease payments with Stripe, iOS, Android, India payment rails, and a roadmap for Finternet-inspired cross-border payments and identity management.
- Built Shyftly, a live workforce-management SaaS platform for small-business owners, managers, and employees, covering multi-store ownership, manager delegation, employee onboarding, shift templates/generation, claims/swaps, availability/time-off approvals, coverage analytics, messaging, reminders, and subscription monetization. Website: shyftly.diggt.cloud.
- Architected Shyftly's Flutter and AWS serverless delivery with Node.js/TypeScript Lambda functions, DynamoDB single-table and GSI access patterns, S3 profile assets, EventBridge automation, SES/APNs/FCM notifications, Secrets Manager, Stripe/Apple billing, and product-lifecycle CRM sync.
- Preparing Shyftly beta onboarding with a 100-location QSR owner, using the shared Diggt infrastructure for AWS, SES, end-user messaging, notifications, database, Lambda, and subscription workflows.
- Built Habit, a live mobile-first habit and goals product with 61 users, supporting tracking, streaks, tasks, reminders, completion analytics, and weekly performance feedback with minimal network dependency. Website: habit.diggt.cloud.
- Architected Habit's local-first Flutter/SQLite app where 99% of operations run on-device, with encrypted per-user backup/restore, JWT auth/sync, timezone-aware reminders, EventBridge/Lambda/SES weekly summaries, opt-in analytics, and Apple/Google subscription tracking.
- Led enterprise alternatives data strategy and modernization of AI-native data engineering platforms across a $500B AUM asset-services data estate.
- Led the Alternatives Data team across three time zones, owning architecture and strategy for approximately 1,500 funds, six accounting systems, and approximately 2,000 product/client-facing operational users.
- Defined enterprise Alternatives Data Strategy aligned to compliance, platform modernization, reporting quality, and business priorities.
- Standardized ingestion from six accounting systems into a unified schema and operating model, reducing custom-reporting cycle time by 40% from requirements gathering through final client submission.
- Introduced Eliza and MCP server workflows for governed data queries and report building, reducing business-user time-to-insight by 70%.
- Designed audit-controlled reporting pipelines where every action and report change was captured for traceability.
- Won BNY AI Hackathon with an anomaly-detection solution that was adopted into production monitoring.
- Led engineering for an enterprise media SaaS platform serving professional creative workflows and major media organizations including CNN and Disney.
- Scaled the globally distributed engineering organization from 2 to 45 while supporting 2 enterprise customers and approximately 3,000 platform users across US and UK workflows.
- Re-architected the initial MVP into a microservices-based enterprise media platform with approximately 15 services managing approximately 800TB of data and multi-region transcoding clusters that auto-tuned capacity based on request volume.
- Reduced infrastructure costs by 80% by replacing an external cloud transcoding dependency with an in-house cloud-based transcoding service, plus architecture optimization and vendor renegotiation.
- Maintained 99.99% uptime for more than a year across pipelines and infrastructure supporting global customers and demanding production workloads.
- Integrated platform workflows with Adobe, Final Cut Pro, and DaVinci Resolve, reducing editorial turnaround time by 50%.
- Moved release operations from ad hoc delivery to structured, certified, well-tested biweekly releases.
- Helped drive $8M ARR through camera-to-cloud partnerships, workflow integrations, and strategic ecosystem delivery.
- Supported NFT creation and submission to Ethereum using smart contracts as part of platform expansion.
- Led enterprise engineering modernization across private cloud, analytics, business-process digitization, and DevOps automation.
- Built private cloud infrastructure using VMware and Docker, reducing monthly hosting spend from approximately $30K to approximately $8K by moving dev/QA nodes to private infrastructure while keeping production cloud-hosted.
- Designed distributed analytics platforms using Hadoop, Spark, HDFS, and Elasticsearch.
- Increased deployment velocity 5x, enabling weekly releases across multiple client platforms through SDLC standardization, Scrum/agile practices, CI/CD automation, and disciplined QA/recovery workflows.
- Led a 40-member engineering organization across 20% onsite and 80% offshore delivery, supporting approximately 80 clients.
- Digitized business processes across HR, sales, engineering, marketing, timesheets, CRM, and service delivery operations.
- Owned OpEx/CapEx planning, third-party vendor relationships, negotiations, and business/technology alignment across modernization programs.
- Led engineering strategy and platform development for global trading analytics, regulatory reporting, data platforms, entitlement systems, and high-volume financial workflows.
- Managed a 46-node Sybase IQ cluster in multiplex mode with three live-live clusters for business and regional isolation across fixed income, New York, and London.
- Supported high-volume financial data platforms processing roughly 2TB of new data daily across trading analytics, regulatory reporting, warehouse, and entitlement workflows.
- Reduced MemSQL operational and support overhead by 75% over six months through workflow redesign, platform tooling, automation, issue-pattern analysis, and onboarding new strats teams to self-service patterns.
- Built enterprise data platforms using Hadoop, Spark, HDFS, MemSQL, Sybase IQ, Sybase ASE, DB2, Elasticsearch, MongoDB, Kafka, Java, and Scala.
- Designed and architected GSAM Entitlement Engine, a generic scalable entitlement framework applying logical access rules across HDFS, Sybase IQ, Sybase ASE, MemSQL, Elasticsearch, MongoDB, and COPTR access workflows.
- Built database metrics and query analytics platforms using Spark, HDFS, MemSQL, Scala, Java, Parquet, and time-series database/server metrics to help users diagnose SQL and platform performance.
- Managed MemSQL platform automation, including single-click cluster deployment, in-place upgrades, online migration runbooks, and ownership transition patterns for 16 production clusters; reduced deployment and upgrade cycle time from days to hours.
- Developed data lake ingestion and readiness frameworks for strategic warehouse APIs supporting CSV, SQL, Parquet, Kafka, raw-to-curated processing, and virtual warehouse consumption.
- Re-architected the Swaps Data Repository Gateway for OTC derivative regulatory reporting across Credit, Rates, FX, Equity, and Commodities; supported daily and near-real-time SDR reporting for CFTC, SEC, ESMA, NSD, TK, and Korean requirements.
- Designed DB2 partitioning and workflow isolation for regulatory reporting, reducing database size from 1.7TB to 900GB while improving backup resilience and maintainability.
- Owned core derivatives connectivity and messaging workflows with DTCC, Markitwire, and DerivServ for confirmations, affirmations, settlements, and daily warehouse reconciliation.
- Built federated development models for FpML validation and gateway workflows, reducing Tier A support overhead from a three-person team to one headcount.
- Delivered front-office trading and sales technology including trade-entry tools, institutional sales merchandising, and electronic trading platform enablement.
- Served as Co-COO for Technology Asian Professionals Network, partnering with HCM and technology/business leadership on mentoring, retention, career development, budgeting, and divisional programming.
- Built financial technology systems before joining Goldman Sachs.
Education
Fairleigh Dickinson University
MS, Computer Science
University of Madras
BS, Computer Science Engineering
Technical Skills
AI and LLM Systems: Enterprise AI platforms, LLM infrastructure, RAG, private inference, vLLM, Ollama, LiteLLM, MCP-style tool execution, model routing, evaluation loops, governed memory, policy-aware workflows.
Data Platforms: Spark, Hadoop, HDFS, Sybase IQ, Sybase ASE, MemSQL, DB2, Elasticsearch, MongoDB, Kafka, Parquet, SQL, data lakes, warehouse APIs, entitlement frameworks, lineage, audit controls.
Cloud and Platform Engineering: AWS, EKS, GKE, AKS, k3s, Docker, Kubernetes, Helm, CDK, Cloudflare Workers/Pages, API Gateway, Lambda, DynamoDB, S3, EventBridge, SES, VMware, CI/CD, observability, cost optimization.
Software Engineering: Java, Scala, Node.js, TypeScript, Python, Flutter, SQLite, microservices, serverless architecture, API design, authentication, authorization, RBAC, OIDC/JWT, security hardening.
Leadership: CTO leadership, VP/SVP Engineering, global engineering organizations, distributed teams, operating model design, vendor negotiation, OpEx/CapEx planning, regulated systems delivery, product strategy.