Open to remote Staff / Principal / Lead roles

Ayush Soni

Staff Backend Engineer — Platform & AI Infrastructure

I've spent 10+ years building distributed systems that serve millions of users across e-commerce, travel, and AI/ML products. I love taking monoliths apart, designing resilient microservices, and building self-serve platforms that remove bottlenecks for entire teams. Early adopter of AI-assisted engineering — Claude Code, Cursor, Windsurf, and MCP servers are part of my daily workflow.

10+ yrs
Backend engineering experience across 5 companies
10d → 24h
Cut Salesforce tenant migration time via automation
20M/day
Personalized emails delivered at Priceline
2.5×
Team migration capacity via self-serve tooling

Where I've built

A decade of shipping backend systems at real scale — from SMB storefronts to Fortune 500 cloud migrations.

Senior Member of Technical Staff
Salesforce — Commerce Cloud
Nov 2022 — Present
Worked on the 1P → Hyperforce (Public Cloud) migration for Salesforce Commerce Cloud — contributed to the first successful 1P tenant migration. Helped cut migration time from 10 days to <24 hours via fully automated zero-downtime pipelines orchestrated with Temporal. Built self-serve tools that 2.5×'d team capacity. Modernized backend stack: JDK 11, Terraform, K8s 1.29+, RBAC. Integrated Claude, Cursor, Windsurf, and MCP servers into daily engineering workflows across the team — accelerating code generation, review, and test authoring.
JavaSpring BootDynamoDBTemporalKubernetesTerraformAWSClaudeCursorMCP
Lead Software Engineer
Eureka AI — AI/ML telco analytics
Mar 2021 — Nov 2022
Architected DS2 Feature Store — a platform-as-a-service data science product with 6+ microservices (Feature Store, Metadata Store, Datalens, API gateway). Built the core feature engine driving platform scalability. Shipped the Metadata Store on Elasticsearch + Neo4j with GraphQL for low-latency ML artifact querying. Mentored 4+ engineers.
JavaSpring BootGraphQLElasticsearchNeo4jKafkaKubernetesIstio
Senior Software Engineer II
Priceline — Booking Holdings
Apr 2019 — Mar 2021
Redesigned a monolithic email system into microservices serving ~20M personalized emails/day. Built RESTful APIs (Java/Spring/Kafka/BigQuery) powering personalized delivery based on user search and booking history. Integrated with Iterable for scalable delivery.
JavaSpring BootKafkaGCPBigQueryGKEGraphQL
Software Developer
Pepperfry — E-commerce
May 2018 — Apr 2019
Broke down legacy monoliths into microservices for checkout, refund, and wallet domains. Built a product recommendation engine on Neo4j and a data warehouse on AWS Redshift for BI reporting across products, orders, and customers.
JavaSpringNeo4jRedshiftKinesisLambdaRedis
Software Engineer
Justdial
Oct 2015 — May 2018
Built backend REST APIs for Justdial Omni — a cloud-hosted SMB platform for online orders, inventory, billing, and purchase orders. Enabled real-time inventory tracking with automated distributor stock flows.
JavaSpring MVCHibernateMySQLMongoDBSolr

Things I've built in the open

Side projects and reusable libraries — most of my production work lives behind NDAs, but here's what's public.

What I work with

Deep Java/Spring expertise with a full distributed-systems toolbox across two cloud providers and modern AI infrastructure.

Languages & Frameworks

JavaSpring BootSpring MVCSpring DataHibernateGraphQL-Java

Architecture

MicroservicesEvent-drivenDistributed systemsREST APIsGraphQLMonolith migrationHigh availability

Databases

MySQLPostgreSQLDynamoDBMongoDBElasticsearchNeo4jRedshiftBigQueryRedis

Streaming & Messaging

KafkaAWS KinesisAWS SQS/SNS

Cloud

AWSLambdaEC2S3RDSGCPBigQueryGKEFirestore

Infra & DevOps

DockerKubernetesIstioHelmTerraformTemporalJenkinsHarness

Observability

OpenTelemetryNew RelicSplunkELKGlowrootSonarQube

AI Engineering

Claude CodeCursorMCP ServersWindsurfFeature StoresML Metadata

Building something interesting?

I'm open to remote Staff, Principal, Lead, or Architect roles at companies building at real scale — platforms, infrastructure, or AI-native products. Happy to work with any timezone.