Nagarjun Mallesh.

I build backend, cloud, automation, and applied AI systems from messy workflows to production-grade software.

Seeking Applied AI FDE / backend / cloud automation roles

View systemsResume
Nagarjun Mallesh

Based in the US

New York, NY. Open to relocation.

Enterprise automationBackend/cloud systemsApplied AI/RAGProduct engineering
Available for thoughtful engineering workEnterprise automationApplied AIBackendCloud systems

Live systems

Systems I have built.

Three live systems that extend the same engineering pattern: clarify the workflow, define the system boundary, and make the result usable enough for people to trust.

01

Interactive systems education

Visual Learning

Visit live project
Problem

System design topics are often explained with static diagrams that hide runtime behavior, failure modes, and the tradeoffs engineers actually debug.

Built

A client-side simulator with interactive modules across databases, Kafka, TCP, rate limiting, transformer inference, sharding, and scaling patterns.

Technical depth

Typed simulation engines, state-driven controls, SVG visualizations, live metrics, and module-level architecture for adding new system internals.

Why it matters

Turns abstract infrastructure concepts into inspectable systems for engineers who learn better by watching behavior unfold.

  • React
  • TypeScript
  • Vite
  • Tailwind
  • Framer Motion
  • SVG

02

Resume intelligence platform

Prism Pro

Visit live project
Problem

Resume tools either give generic advice or fabricate improvements candidates cannot defend in interviews.

Built

A FastAPI and Next.js platform for resume parsing, recruiter-style evaluation, JD tailoring, versioning, preview, and country-aware PDF export.

Technical depth

Schema-constrained LLM responses, hallucination guards, ATS simulation, Supabase auth/storage, credit accounting, and Playwright PDF rendering.

Why it matters

Helps candidates improve documents with evidence they can explain, not model-generated polish that collapses under scrutiny.

  • FastAPI
  • SQLAlchemy
  • Pydantic
  • OpenAI
  • Next.js
  • Supabase

03

AI outreach assistant + Chrome extension

Aletheia

Visit live project
Problem

AI outreach often sounds generic, unsafe, or disconnected from real profile context.

Built

A Chrome MV3 extension and Next.js app that generate grounded LinkedIn notes, cold emails, and InMails from profile context, resume data, and user intent.

Technical depth

Prompt guardrails, sanitization, AI-fingerprint stripping, feedback loops, style learning, rate limits, Supabase auth, and Anthropic Claude integration.

Why it matters

Keeps personalization grounded in visible context while giving users faster drafts they can still review and own.

  • Next.js
  • TypeScript
  • Chrome MV3
  • Supabase
  • Claude
  • Zod

Technical stack

Tools and systems I work with.

Languages

  • Python
  • TypeScript
  • JavaScript
  • Java
  • Go
  • Bash
  • PowerShell

Backend

  • FastAPI
  • Node.js
  • Express
  • Spring Boot
  • REST APIs
  • microservices

Cloud infra

  • AWS Lambda
  • EC2
  • S3
  • API Gateway
  • RDS
  • CloudWatch
  • EventBridge
  • Terraform
  • Docker

Data

  • PostgreSQL
  • MongoDB
  • Redis
  • DynamoDB
  • Kafka

AI systems

  • RAG
  • AWS Bedrock
  • Claude
  • Hugging Face
  • FAISS
  • Sentence Transformers
  • RAGAS

Engineering spectrum

Enterprise systems to applied AI.

The throughline is not a single framework. It is building reliable software around messy operational workflows, infrastructure constraints, and emerging AI capabilities.

01

Enterprise Provisioning & Automation

Automated BIOS configuration, OS deployment, and HP hardware provisioning workflows, reducing manual setup from days to minutes and supporting enterprise device rollout at scale.

  • PowerShell
  • React
  • Node.js
  • BIOS config
  • OS deployment
  • 100K+ machines

02

Cloud Backend Infrastructure

Built hybrid AWS/GCP storage flows, event-driven processing, RDS-backed schemas, containerized Node.js services, and CloudWatch observability for production backend systems.

  • AWS/GCP
  • Lambda
  • EventBridge
  • RDS Multi-AZ
  • Docker/ECR
  • Terraform

03

Applied AI & RAG Systems

Designed RAG and LLM workflows with AWS Bedrock, hybrid retrieval, RAGAS evaluation, prompt templates, and text normalization for client-facing business workflows.

  • Bedrock
  • Claude
  • LangChain
  • RAGAS
  • FAISS
  • Sentence Transformers

Engineering arc

Useful before impressive. Operable before clever.

(02)

I moved toward applied AI for the same reason I kept building automation: to reduce manual error, make complex workflows easier to operate, and turn emerging capability into software teams can trust in production.

2025

Applied AI and RAG systems

Designed Bedrock RAG pipelines, hybrid retrieval, RAGAS evaluation, prompt templates, and LLM text normalization workflows at ML Technologies.

2024

Cloud backend and infrastructure

Built AWS/GCP storage services, RDS Multi-AZ schemas, Lambda/EventBridge processing, Dockerized Node.js services, and CloudWatch observability.

2019-2022

Enterprise automation and services

Automated BIOS configuration, OS deployment, provisioning dashboards, and Spring Boot/Kafka service migration work across enterprise environments.

Writing and learning

Writing to clarify what I am learning.

I write to clarify what I am learning and share practical notes with other engineers. My writing explores backend systems, AI workflows, automation, and the implementation tradeoffs I encounter while building.

MediumSubstack
Substack / Feb 18, 2026The Quiet Revolution in Document ExtractionMedium / Nov 17, 2025From Token Tears to Triumphant CodeSubstack / Jan 16, 2026Is My Spotify Web Player Broken?

Contact

Available for applied AI FDE, full-stack systems, backend, and cloud automation roles.

Best fit: teams that need someone to translate ambiguous technical problems into reliable APIs, applied AI workflows, product surfaces, and infrastructure automation.

EmailLinkedInGitHub