Rajan Kalwar · Senior Software Engineer
I build reliable cloud and AI systems.
Senior software engineer building production systems on AWS—distributed backends, event-driven architectures, containerized workloads, and GenAI applications—engineered for reliability, security, observability, and cost.
AWS · Python · TypeScript · Serverless · Containers · GenAI
Experience
- Describe what you built or led here, leading with an action verb.
- Add a second impact-focused bullet in the same voice as your other roles.
- Note a third system or architecture contribution.
- Built secure and scalable LLM-powered applications and RAG pipelines using LangChain, LangGraph, and LlamaIndex, with a focus on efficient token usage and reliable response generation.
- Designed and deployed cost-efficient serverless architectures on AWS using Lambda, DynamoDB, Cognito, API Gateway, and CloudFormation.
- Built and operated containerized workloads on AWS ECS with private VPC networking, ECR, and least-privilege IAM permissions.
- Designed scalable SQL and NoSQL data models using PostgreSQL and DynamoDB for high-performance application workloads.
- Developed distributed microservices using NestJS, Redis, and AWS App Runner.
- Designed and developed GraphQL APIs using AWS AppSync, Lambda, and DynamoDB.
- Built high-throughput DynamoDB workloads using single-table design, optimized access patterns, and scalable data modeling.
- Designed and maintained AWS infrastructure supporting frontend and backend application workloads.
- Provisioned and managed cloud resources for reliable and scalable application delivery.
- Built managed API-as-a-Service solutions using AWS API Gateway, including API keys, usage plans, throttling, and rate limiting.
- Contributed to the design and implementation of AWS infrastructure across multiple projects.
- Provisioned and maintained cloud resources supporting REST and GraphQL APIs.
- Built and maintained CI/CD pipelines, automated deployments, and managed infrastructure as code using AWS CloudFormation.
Technical skills
Languages and frameworks
- Python: FastAPI, Flask, Pydantic, AWS Lambda Powertools
- TypeScript / JavaScript: NestJS, Fastify, Feathers, Node.js
- Go: Fiber, Buffalo
- API design: REST, GraphQL, Server-Sent Events
AWS compute and APIs
- Serverless: Lambda, API Gateway, AppSync, Cognito
- Containers: ECS, Fargate, App Runner, ECR, Application Load Balancer
- Compute: EC2, Auto Scaling, Lambda Function URLs
- Content delivery: S3, CloudFront
Data and event systems
- Databases: DynamoDB, PostgreSQL, RDS, Redis, MongoDB, ArangoDB
- Messaging: SQS, SNS, EventBridge, DynamoDB Streams
- Orchestration: Step Functions, event-driven consumers, asynchronous fan-out
- Data modeling: DynamoDB single-table patterns, SQL and NoSQL access patterns
Networking, identity, and security
- Networking: VPC, public and private subnets, NAT, security groups, private DNS
- Identity: IAM, Cognito, OAuth, SAML, OIDC roles
- API controls: Usage plans, rate limits, API keys, field-level authorization
- Security practice: Least-privilege permissions, private workloads, secure artifact delivery
Production GenAI
- Frameworks: LangChain, LangGraph, LlamaIndex
- Retrieval: RAG, vector retrieval, document ingestion, context selection
- Delivery: Model-provider integration, response streaming, tool calling
- System concerns: Secure document access, token usage, API boundaries, operational cost
Infrastructure and reliability
- Infrastructure as code: CloudFormation, Serverless Framework
- Delivery: Docker, ECR, CI/CD pipelines, staged deployments
- Reliability: Dead-letter queues, partial batch failures, retries, idempotent consumers
- Supporting cloud: GCP App Engine, Cloud Run, Firebase, Firestore, Pub/Sub
Recent writing
AWS
Amazon S3 storage tiers
Choosing storage classes around access patterns and cost.
AWS
CloudFormation parameters
Making infrastructure templates reusable and explicit.
Tooling
Pre-commit hooks with Husky
Moving quality checks earlier in the development loop.
Featured case studies
Client → API Gateway → Lambda → DynamoDB
API access controls with native AWS services
Designed a deployable access-control system using API Gateway usage plans for tier-specific throttles and quotas. A Lambda management API handles key and plan lifecycles, while DynamoDB stores metadata without retaining raw credentials.
- Isolated administrative operations from protected workloads
- Made propagation delays and non-transactional boundaries explicit
HTTP API → Lambda → SQS → SNS → Consumers
Resilient asynchronous fan-out
Engineered an event pipeline that buffers HTTP requests in SQS, processes them asynchronously with Lambda, and distributes results to independent notification and audit consumers through SNS.
- Decoupled request handling through durable queueing
- Retried only failed batch items and routed exhausted messages to a dead-letter queue
- Scoped producers and processors to their minimum required permissions
Internet → ALB → Private subnets → ECS
Private ECS services with explicit capacity trade-offs
Built comparable Fargate and EC2-backed ECS services behind an Application Load Balancer, keeping application workloads in private subnets while making the capacity-ownership trade-offs directly comparable.
- Kept tasks and container instances off the public internet
- Separated networking, capacity, image, and service concerns
- Sequenced infrastructure and image delivery before service rollout
Have a cloud or AI system to build?
I am interested in senior engineering work involving AWS architecture, distributed backends, platform reliability, and production GenAI.