Open to New-Grad SWE / AI Roles

New-grad software engineer building reliable products across backend, mobile, and AI systems.

I am currently building Fleetrac, an agentic-governance platform for monitoring AI systems, detecting incidents, collecting evidence, assigning ownership, and supporting human review and remediation.

View selected workView resume

San Francisco Bay Area · Full-Stack · Applied AI · AI Systems & Agentic Governance · AI Product Design

Portrait of Shulabh Bhattarai

Shulabh Bhattarai

SWE · Applied AI · Agentic Governance

B.A. in Computer Science

Building

Fleetrac - observability-driven governance for a fleet of AI agents

iOS + Android

Cross-platform MindMitra release

Now available on:

3

Engineers led as a student developer to build MindMitra

2

Ongoing research projects on Bio-Agro weapons

Work

Selected engineering projects

Selected engineering projects and case studies.

MindMitra app screenshot 1 of 4
MindMitra app screenshot 2 of 4
MindMitra app screenshot 3 of 4
MindMitra app screenshot 4 of 4

Mobile + Backend

MindMitra

Cognitive-wellness app (iOS + Android)

MindMitra (formerly CognizenX) is a research-backed cognitive-wellness app built with DePauw Neuroscience. I led a three-engineer team in taking over the shipped React Native product, moving AI and data workflows to an independently deployed backend, and delivering updated iOS and Android builds with quiz analytics and spaced-repetition tracking.

Role
Software Engineer – MindMitra
Core challenge
Take over a live v1 product under its prior CognizenX branding, refactor architecture for a backend-first AI stack, expand to Android, and support neuroscience research workflows without disrupting existing users.
  • Led 3 engineers modernizing a shipped cognitive-wellness app—delivered v2 on iOS and Android.
  • Applied a research-backed approach with DePauw Neuroscience to implement analytics-driven cognitive monitoring and spaced-repetition retention.
  • 25+ endpoint backend on Vercel/MongoDB Atlas with GPT-4 content pipeline and 57 automated tests.
  • React Native
  • Node.js
  • Express
  • MongoDB Atlas
  • GPT-4
  • BullMQ

Previously released as CognizenX on the App Store; the product is now branded MindMitra.

Case study
Now available on:
MediVise screenshot 1 of 2
MediVise screenshot 2 of 2

Full-Stack + Applied AI

MediVise

Healthcare document assistant (senior capstone)

Built a full-stack healthcare accessibility web app that helps patients review complex medical documents, manage medications and appointments, and chat with a grounded AI assistant—designed as a safety-bounded capstone prototype, not a clinical product.

Role
Capstone Developer
Core challenge
Make medical document information more accessible while keeping AI outputs bounded, source-grounded, and clearly non-diagnostic within a capstone scope.
  • Phi-4-mini RAG chat with hybrid retrieval over medical documents.
  • Auth, document CRUD, and med/appointment workflows—bounded, non-diagnostic AI.
  • React
  • FastAPI
  • PostgreSQL (Supabase)
  • phi-4-mini RAG

Educational prototype; not intended for diagnosis or medical advice.

Case study
StudyAI app screenshot 1 of 6
StudyAI app screenshot 2 of 6
StudyAI app screenshot 3 of 6
StudyAI app screenshot 4 of 6
StudyAI app screenshot 5 of 6
StudyAI app screenshot 6 of 6

Mobile + ML inference

StudyAI

AI-native iOS study companion for live lecture capture

Built an iOS study companion that turns in-class live notes and voice capture into structured summaries and machine-generated quizzes, powered by Meta's BART seq2seq model on a FastAPI inference layer with offline-first Firestore and Core Data sync.

Role
Solo Developer
Core challenge
Close the gap between live lecture delivery and effective self-review by converting unstructured in-class input into retention-ready summaries and assessments on a mobile-first, latency-sensitive path.
  • Live lecture capture → BART summaries and auto-generated quizzes.
  • Offline-first iOS (Firestore + Core Data); p95 inference under 400ms.
  • SwiftUI
  • FastAPI
  • BART (facebook/bart-large-cnn)
  • Firebase
Case studyVideo demo
View all work →

Experience

Engineering experience

Full experience

Software Engineer – MindMitra

DePauw Neuroscience Department

Greencastle, IN

  • Took technical ownership of a shipped React Native cognitive-wellness app, leading a three-engineer team in moving AI and data workflows from the mobile client to an independently deployed backend while delivering updated iOS and Android builds.
  • Architected and deployed a 25+ endpoint Node.js/Express API on Vercel with MongoDB Atlas, centralizing authentication, AI quiz content generation, personalized learning state, and quiz performance telemetry.
  • Built an automated demand-aware GPT-4 content pipeline that identifies question-bank gaps across 43 subdomains and applies embedding-based semantic deduplication before persistence.
  • Migrated AI operations server-side after exposed client-side credentials; added session authentication, Joi validation, tiered rate limits, and 57 automated tests across 18 Jest/Supertest suites.
  • React Native
  • Node.js
  • MongoDB Atlas
  • GPT-4
  • BullMQ
  • Redis
  • Expo EAS
  • Jest

Software Engineering Intern

Togglecorp

Remote

  • Developed a React-based analytics dashboard for a UK client, visualizing consumer trends across 200K+ users in 5 regions, with lazy loading, Chart.js visualizations, and filtering by spending tiers and regional segments.
  • Implemented backend APIs and database queries using Django REST and PostgreSQL, optimizing slow client-reported queries through indexing and query tuning, and supporting staging deployments alongside senior engineers.
  • Built exploratory customer segmentation and sales forecasting prototypes using k-means clustering and a lightweight MLP, visualizing results to inform ad targeting, recommendations, and new product sales planning.
  • React
  • Django REST
  • PostgreSQL
  • Chart.js
  • K-means
  • MLP

Engineering focus

Where I contribute

Backend & systems

APIs, data modeling, query optimization, authentication, queues, caching, and service integrations.

Product development

React and React Native applications from core features through testing, release, and iteration.

AI Systems & Agentic Governance

AI integrations, runtime observability, incident detection, evidence collection, guardrails, and human-review workflows.

Data & analytics

Dashboards, retrieval pipelines, segmentation prototypes, and ETL-style data workflows.

AI-assisted development

  • Cursor
  • Codex
  • Claude Code

Agentic AI systems

  • Claude Agent SDK
  • OpenAI Agents SDK
  • MCP
  • LangGraph

Observability and governance

  • OpenTelemetry
  • Langfuse
  • Open Policy Agent

Core stack

Languages & product

  • Python
  • TypeScript
  • React
  • React Native

Backend & data

  • FastAPI
  • Node.js
  • PostgreSQL
  • Redis
  • MongoDB
  • pgvector

Delivery & infra

  • Docker
  • GitHub Actions
  • AWS EC2

About

Building beyond the prototype

I build software where product ownership, backend reliability, and responsible AI systems meet. I am especially interested in taking products beyond demos—making them secure, observable, maintainable, and useful in real workflows.

Through Fleetrac, I am exploring how agentic systems can be monitored and governed through runtime evidence, incident detection, accountable ownership, human review, and remediation.

Read more about my approach →

Open to new-grad software engineering roles

Backend, full-stack, product engineering, and applied AI — where I can contribute to shipped systems.

Get in touchView resume

IBM SkillsBuild certifications: IBM AI Fundamentals · IBM Quantum Enigma