Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Software engineer with 5+ years designing and shipping scalable backend systems, asynchronous processing pipelines, and event-driven architectures across cloud-native environments on AWS and Kubernetes. Proficient in Python (FastAPI, Django, Flask), Java (Spring Boot), PostgreSQL, and Kafka, with a track record integrating ML inference and RAG pipelines into high-volume production systems. Open-source author of pgsemantic (a Python library for semantic search over Postgres) and builder of production-grade security, CI/CD, monitoring, and automated data workflows. Experience delivering real-time ingestion, optimized APIs, and ML-based ticket classification and risk scoring at scale.…Software engineer with 5+ years designing and shipping scalable backend systems, asynchronous processing pipelines, and event-driven architectures across cloud-native environments on AWS and Kubernetes. Proficient in Python (FastAPI, Django, Flask), Java (Spring Boot), PostgreSQL, and Kafka, with a track record integrating ML inference and RAG pipelines into high-volume production systems. Open-source author of pgsemantic (a Python library for semantic search over Postgres) and builder of production-grade security, CI/CD, monitoring, and automated data workflows. Experience delivering real-time ingestion, optimized APIs, and ML-based ticket classification and risk scoring at scale.

Sairam Varmabudh araj u

Back-End Developer, DevOps Developer, Architect, +1





Software engineer with 5+ years designing and shipping scalable backend systems, asynchronous processing pipelines, and event-driven architectures across cloud-native environments on AWS and Kubernetes. Proficient in Python (FastAPI, Django, Flask), Java (Spring Boot), PostgreSQL, and Kafka, with a track record integrating ML inference and RAG pipelines into high-volume production systems. Open-source author of pgsemantic (a Python library for semantic search over Postgres) and builder of production-grade security, CI/CD, monitoring, and automated data workflows. Experience delivering real-time ingestion, optimized APIs, and ML-based ticket classification and risk scoring at scale.…Software engineer with 5+ years designing and shipping scalable backend systems, asynchronous processing pipelines, and event-driven architectures across cloud-native environments on AWS and Kubernetes. Proficient in Python (FastAPI, Django, Flask), Java (Spring Boot), PostgreSQL, and Kafka, with a track record integrating ML inference and RAG pipelines into high-volume production systems. Open-source author of pgsemantic (a Python library for semantic search over Postgres) and builder of production-grade security, CI/CD, monitoring, and automated data workflows. Experience delivering real-time ingestion, optimized APIs, and ML-based ticket classification and risk scoring at scale.

Available to hire

Software engineer with 5+ years designing and shipping scalable backend systems, asynchronous processing pipelines, and event-driven architectures across cloud-native environments on AWS and Kubernetes. Proficient in Python (FastAPI, Django, Flask), Java (Spring Boot), PostgreSQL, and Kafka, with a track record integrating ML inference and RAG pipelines into high-volume production systems.

Open-source author of pgsemantic (a Python library for semantic search over Postgres) and builder of production-grade security, CI/CD, monitoring, and automated data workflows. Experience delivering real-time ingestion, optimized APIs, and ML-based ticket classification and risk scoring at scale.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Beginner

Language

Javanese

Advanced

Work Experience

Software Engineer at Uber USA

February 1, 2026 - Present

Built real-time event ingestion pipelines using Kafka and Python to process rides and support telemetry, achieving sub-100ms data freshness for dashboards used by 50+ support agents. Redesigned the query layer across MySQL and MongoDB via composite indexing and schema tuning, reducing p95 retrieval latency from 3.2s to 1.4s for customer support backend services. Architected REST APIs in Node.js and Express with role-based access control, enabling secure data exchange across customer support microservices and reducing auth-related defects by 3 per release. Integrated ML-based ticket classification into the support routing pipeline using FastAPI and scikit-learn, automating triage for 12K+ weekly tickets and reducing average first-response time by 35%. Architected CI/CD pipelines using GitHub Actions and AWS CodePipeline for containerized services on EKS, cutting deployment lead time by 6 hours per sprint and eliminating manual release steps.

Software Developer at Northeastern University Washington, USA

March 1, 2025 - December 31, 2025

Built AI-powered grant eligibility workflows using RAG with OpenAI GPT and LangChain, served via FastAPI with scikit-learn classification models, processing 300+ records per cycle with 92% accuracy. Implemented event-driven enrollment and grant processing pipelines using Apache Kafka and Spring Boot on AWS EKS, designing topic partitioning to sustain 5K+ events/hour with zero message loss. Refactored legacy auth modules to OAuth 2.0 and JWT on Kubernetes, reducing production defects from 8 to 2 per release cycle and cutting auth-related incident response by 60%. Deployed Prometheus and Grafana monitoring across 4 microservices, enabling proactive alerting that reduced undetected service degradation by 70% and cut mean time to recovery from 45 to 12 minutes.

Software Developer at University of Florida Florida, USA

May 1, 2024 - February 28, 2025

Developed REST APIs using FastAPI and PostgreSQL for research data access, reducing average query response time by 35% through indexing and query plan optimization. Automated recurring ETL workflows using Apache Airflow and Python, cutting report generation time from 6 hours to 2 hours weekly across student and administrative datasets. Deployed microservices on AWS EKS using Docker and Kubernetes, configuring health checks and autoscaling to maintain 99.5% uptime across internal research portals. Integrated OAuth 2.0 authentication and set up GitHub Actions CI/CD pipelines for automated testing and zero-touch deployments across dev and staging environments.

Software Developer/Assistant Systems Engineer at Tata Consultancy Services India

October 1, 2020 - April 30, 2022

Built backend transaction services using Flask, SQLAlchemy, and Oracle database integration across 4 financial source systems, processing 300K+ daily records with 99.9% settlement accuracy for production reporting. Orchestrated 130+ daily settlement jobs using Celery and RabbitMQ, reducing batch runtime from 2 hours to 55 minutes and improving SLA adherence by 30%. Built an XGBoost risk scoring pipeline processing 8K+ daily financial transactions, flagging high-risk entries with 87% precision and reducing manual review workload by 45% for the operations team. Engineered PySpark and pandas data transformation pipelines consolidating 800K+ weekly records from 6 upstream systems, cutting reconciliation processing time by 50%. Automated monthly data ingestion using Azure Functions, eliminating 3 full days of manual processing per cycle and improving dataset delivery reliability across 4 client accounts.

Software Developer at Airbnb India

May 1, 2019 - September 30, 2020

Connected booking workflows to ML recommendation models by building a Spring Boot to Flask inference shim on AWS EC2, handling 4.5K calls/day and improving p95 latency from 1.2s to 0.8s. Orchestrated nightly Airflow pipelines syncing PostgreSQL to Snowflake, preparing 2–3 pricing datasets daily and eliminating manual analyst data pulls. Automated CI/CD pipelines with Jenkins, reducing manual deployment efforts by 40%, and implemented containerization with Docker and Kubernetes for zero-downtime deployments. Expanded integration test coverage by 40+ cases through schema contract alignment with PM, DS, and QA partners, reducing cross-team integration bugs by 30%. Prototyped a semantic search engine using TF-IDF and Elasticsearch, indexing 1.2K property descriptions and reducing onboarding lookup time to under 450ms (p50).