0%
INITIALISING SYSTEMS
EXPERIENCE WORK WRITING SKILLS CONTACT
LAT 37.7749 LNG -122.4194 SAN FRANCISCO, CA SYS.ONLINE VER 2.0
Sudheendra Sabnavisu

SUDHEENDRA SABNAVISU.

Backend Engineer · Distributed Systems · GenAI

4+ years building scalable APIs, microservices, and Kafka-driven systems in Java, Spring Boot, Python, and FastAPI — plus GenAI features that ship to production and stay predictable under load.

AVAILABLE FOR OPPORTUNITIES

I design systems that stay calm under pressure.

From 1,000-QPS REST APIs to event-driven pipelines and RAG platforms, I focus on predictable latency, clear observability, and architectures teams can trust without second-guessing.

01 — Identity

The philosophy.

I'm drawn to messy systems — not for the chaos, but for the chance to find the signal inside the noise. From migrating workloads to AWS at Salesforce to building Kafka-backed event pipelines and citation-grounded RAG services, I stitch together APIs, data, and LLMs so complex systems feel simple to the people who depend on them.

4+
Years Engineering
1K QPS
Peak API Throughput
170M+
Data Points Analyzed
↓ DOWNLOAD RESUME
02 — Career

Experience.

Banner Life Family of Companies
Sep 2025 — Present · Remote
Software Engineer (Contract)
  • Owned end-to-end design, development, and production deployment of 2 high-throughput REST APIs at peak loads of 1,000 QPS, synchronizing lead, policy, and application-status data across CRM and underwriting platforms.
  • Secured 100% of external API traffic with OAuth 2.0 client-credentials flow, token-based authentication, and RBAC — alongside structured logging, payload validation, and automated recovery with up to 5 retries.
  • Architected a Python-based generative AI platform for an enterprise learning and compliance system, launching RAG-powered assistance, automated assessment generation, and AI-driven course summarization.
  • Built the RAG pipeline on LLM APIs, embeddings, semantic search, metadata filtering, and citation-backed responses — cutting learner content-discovery time by 60% and grounding answers in authorized material.
  • Automated quizzes, answer keys, and difficulty-tagged assessments, reducing content-authoring time by over 80% and letting instructors produce a 10-question assessment in under 5 minutes with human review.
PythonREST APIsOAuth 2.0RBACRAGEmbeddingsLLM APIs
Physis Investment
Jan 2025 — Apr 2025 · Boston, MA
Software Engineer
  • Built and deployed FastAPI services for ImpactChat, a proprietary ESG research assistant using RAG, PostgreSQL, and PGVector to analyze 170M+ data points across 14K+ companies and 335K+ funds.
  • Designed a Kafka-backed feedback pipeline capturing thumbs-up/down ratings asynchronously, enabling analytics, model evaluation, and end-to-end traceability across prompts, responses, and inference metadata.
  • Developed streaming APIs and D3.js visualizations for portfolio analysis, industry benchmarking, and historical trends, delivering ESG insights through investor-facing dashboards.
  • Automated build, test, and deploy to auto-scaled AWS EC2 with GitHub Actions and Terraform, cutting production deployment time by 65% and enabling weekly releases.
PythonFastAPIPGVectorKafkaAWSTerraformD3.js
Salesforce
Jul 2021 — Aug 2023 · Bengaluru, India
Software Engineer
  • Migrated application workloads to AWS using Packer-based immutable images, VPC peering, private networking, and secure database connectivity — cutting provisioning time by 70% and eliminating configuration drift.
  • Built a Kafka-based onboarding event pipeline consuming new-hire events and triggering SendGrid welcome emails, IT account provisioning, compliance training, and manager notifications — decoupling 4 downstream workflows with retry and replay support.
  • Designed and optimized backend services and automation pipelines (Apex, JavaScript, batch processing), reducing system latency by 20% and improving data accuracy by 35%.
  • Improved query performance through SQL optimization, achieving 25% faster execution on high-volume datasets.
AWSPackerKafkaApexSQLEvent-Driven
Resideo
Jan 2021 — Jun 2021 · Bengaluru, India
Software Engineering Intern
  • Built an Elasticsearch indexing pipeline for thermostat telemetry and application logs — searchable by device ID, timestamp, temperature, setpoint, HVAC state, and error code — plus Kibana dashboards that cut production triage time by 40%.
  • Developed a modular Java application generating customizable JSON payloads with object-oriented design patterns, enabling reuse across testing teams and improving testing efficiency by 30%.
  • Authored API documentation to OpenAPI specification with Swagger, standardizing interface contracts across services.
JavaElasticsearchKibanaOpenAPISwagger
03 — Education

Education.

Northeastern University
Sep 2023 — Jul 2025 · Boston, MA
Master of Science, Computer Software Engineering
Graduate coursework across distributed systems, cloud computing, web design, and program structures — the foundation behind the services and pipelines listed above.
Amrita School of Engineering
2017 — 2021 · Amritapuri, Kerala, India
Bachelor of Technology, Computer Science
Undergraduate foundation in data structures, algorithms, operating systems, and networks — where the systems thinking started.
04 — Recent Builds

Systems I'm proud
to have shipped.

01

BookVerse

A cloud-native RAG service pairing FastAPI, an LLM, and a vector store to answer questions over large book and document collections with citation-backed responses.

FastAPIRAGVector SearchAWS
REPOSITORY →
02

E-Learning Platform

A course delivery platform with role-based access, assessment workflows, and AI-assisted summarization — built around a JAVA backend and a relational content model.

JavaSpringRBACLLM
REPOSITORY →
03

YOLO Object Detection on AWS

A containerized YOLO inference service deployed to AWS, exposing detection over a REST endpoint with reproducible Docker builds and horizontal scaling.

PythonYOLODockerAWS
REPOSITORY →
04

Cloud-Native Setup on GCP

Infrastructure-as-code for a full cloud environment — networking, managed instances, and CI/CD wiring — provisioned from scratch with Terraform and Packer images.

GCPTerraformPackerCI/CD
REPOSITORY →
05

ETag Caching with Redis

HTTP ETag generation and conditional-request handling backed by Redis, cutting redundant payload transfer by returning 304s when resource state is unchanged.

RedisHTTP CachingREST APIs
REPOSITORY →
06

Onboarding System

An event-driven onboarding pipeline: new-hire events fan out over Kafka to trigger welcome emails, account provisioning, and training assignments, with retry and replay on every consumer.

KafkaEvent-DrivenJavaMicroservices
REPOSITORY →
07

Ekadashi App

A lightweight calendar app that computes and surfaces Ekadashi dates by location, with a simple API and a mobile-first interface.

JavaScriptREST APIs
REPOSITORY →
05 — Writing

Latest posts.

Notes on backend systems, event-driven architecture, and building with LLMs.

SEE ALL POSTS →
Java·Spring Boot·Python·FastAPI·Kafka·AWS·Kubernetes·Docker·Terraform·PostgreSQL·PGVector·Redis·Elasticsearch·RAG·LangChain·Microservices·REST APIs· Java·Spring Boot·Python·FastAPI·Kafka·AWS·Kubernetes·Docker·Terraform·PostgreSQL·PGVector·Redis·Elasticsearch·RAG·LangChain·Microservices·REST APIs·
06 — Skills

Technical orbit.

Languages

JavaPythonJavaScriptSQLBash

Backend

Spring BootFastAPIREST APIsMicroservicesAPI DesignSystem DesignOAuth 2.0 / RBAC

Distributed Systems

KafkaRabbitMQRedisEvent-Driven Architecture

Databases & Data

PostgreSQLPGVectorDynamoDBElasticsearchSQL Query Optimization

Cloud & Infrastructure

AWSGCPAzureDockerKubernetesTerraformPackerGitHub ActionsJenkins

AI & LLMs

RAGLangChainEmbeddingsSemantic SearchMCP
07 — Contact

Let's build
something
together →

Have a backend or GenAI system to build, or a role to fill? I'd love to hear about it. Based in San Francisco, open to remote.

EMAIL +1 857-693-9339 LINKEDIN GITHUB RESUME