Sandeep Erelli
Building AI-native platforms and distributed systems at scale. Pioneered 100M+ workflow orchestrations through open-source Temporal — redefining how enterprises automate critical business processes.
// 100M+ workflows orchestrated
func OrchestrateAIWorkflow(
ctx workflow.Context,
payload WorkflowPayload,
) error {
ao := workflow.ActivityOptions{
StartToCloseTimeout: 30 * time.Minute,
RetryPolicy: &temporal.RetryPolicy{
MaxAttempts: 3,
},
}
ctx = workflow.WithActivityOptions(ctx, ao)
var result AIInferenceResult
err := workflow.ExecuteActivity(
ctx, RunAIInference, payload,
).Get(ctx, &result)
if err != nil {
return err
}
return workflow.ExecuteActivity(
ctx, PublishResults, result,
).Get(ctx, nil)
}
// Durable. Scalable. Production-proven.Engineering at the intersection of AI & Scale
I'm a Principal Software Engineer with 8+ years building distributed systems, AI platforms, and developer tooling that powers millions of users. My work sits at the intersection of systems engineering and applied AI — where reliability, scale, and intelligence converge.
Most notably, I architected and led adoption of open-source Temporal across enterprise workflows, orchestrating over 100 million executions with five-nines reliability. This platform became the backbone of mission-critical processes spanning finance, healthcare, and logistics.
Beyond systems work, I contribute to the AI ecosystem — building AI-native developer tools, experimenting with multi-agent architectures, and mentoring engineers on distributed systems and LLM integration patterns.
100M Workflows. One Open-Source Platform.
Temporal Workflow Orchestration at Scale
Open-source · Production · 100M+ executions
Designed and shipped a multi-tenant workflow orchestration platform built on open-source Temporal that processes over 100 million workflow executions. The platform handles complex, long-running business processes spanning hours to months — with automatic retries, durable state, and zero message loss across distributed environments.
AI Platform Architecture
Designed multi-agent AI orchestration systems integrating Claude, GPT-4, and custom fine-tuned models — enabling intelligent workflow routing and autonomous decision-making within durable Temporal workflows.
Open Source Contributions
Active contributor to the Temporal ecosystem — submitting patches, writing SDK examples, and publishing production patterns adopted by teams at Fortune 500 companies and high-growth startups alike.
Engineering Leadership
Mentored 15+ engineers across distributed systems, Go, and AI integration patterns. Established engineering standards, conducted architecture reviews, and drove cross-functional alignment on platform decisions.
Professional Journey
Principal Software Engineer
- Architected a Temporal-based workflow orchestration platform processing 100M+ executions — spanning financial reconciliation, AI inference pipelines, and data synchronization
- Designed multi-agent AI systems using Claude and GPT-4 for autonomous workflow routing — reducing manual intervention by 78% across business-critical processes
- Built real-time observability stack (Prometheus, Grafana, OpenTelemetry) providing P99 latency visibility across 50+ workflow namespaces
- Led SDK strategy for Go and Python Temporal clients; published patterns adopted by 30+ enterprise teams
- Scaled infrastructure from 1M to 100M+ monthly workflow executions with zero-downtime migrations on AWS/EKS
Senior Software Engineer
- Introduced Temporal to replace fragile cron-job pipelines — eliminating 95% of timeout-related production incidents across data ingestion workflows
- Built developer platform enabling internal teams to author, test, and deploy workflows in Go/Python without infrastructure expertise
- Integrated LLM-powered document classification into durable workflows, processing 2M+ documents/month with 94% accuracy
- Reduced P99 API latency by 60% through async workflow offloading and intelligent request batching patterns
- Mentored 8 engineers on distributed systems patterns; conducted weekly system design sessions
Software Engineer
- Built full-stack SaaS products for 10+ clients spanning fintech, healthcare, and e-commerce using React, Node.js, and TypeScript
- Designed event-driven microservices on AWS with SQS/SNS, reducing system coupling and improving deployment velocity by 3x
- Implemented CI/CD pipelines and infrastructure-as-code practices that became team-wide standards
- First-hand experience with complex async process failures — driving eventual deep investment in workflow orchestration solutions
Technical Skills
Featured Projects
Temporal Workflow Platform
Multi-tenant workflow orchestration platform built on open-source Temporal, processing 100M+ executions. Handles complex, long-running AI inference pipelines, financial reconciliation, and multi-step data transformations with durable state, automatic retries, and full distributed observability.
Serelli Lab
AI-native coding platform with live sandbox environments, Claude-powered code generation, and intelligent pair programming. Built with Next.js, Vercel Sandbox, and Anthropic SDK.
AI Workflow Intelligence
Multi-agent system using Temporal for durable orchestration of LLM inference chains. Routes tasks across Claude, GPT-4, and custom models based on task type, cost, and latency requirements — with full audit trails.
Distributed Observability Platform
End-to-end distributed tracing and metrics for Temporal workflows. Workflow-level SLA tracking, alerting, and business process dashboards across 50+ namespaces in production.
Open to conversations
I'm interested in senior engineering roles, technical advisory, and conversations around distributed systems, AI platforms, and workflow orchestration at scale.