Curriculum vitae

Resume

Content last updated October 2026. Use the button to print this page or save it as a PDF.

Explore my work

Harmanpreet Singh

Software engineer · Atlanta, Georgia

harmansinghuga@gmail.com | github.com/mrhpsingh01 | linkedin.com/in/harmanuga

Summary

Software engineer and ML/AI researcher with production full-stack experience from Stylabs AI Native Studio and Lykan Technologies, and graduate research in machine learning, multimodal sensor intelligence, and distributed systems at the University of Georgia. Developed the FACET framework achieving state-of-the-art activity recognition across 5 CASAS smart homes at 1/6th the inference cost of prior multimodal baselines. Built HomeSense, an end-to-end event-streaming and digital-twin platform orchestrating Kafka, FastAPI, React, and Kubernetes around the thesis architecture.

Experience

University of Georgia, ML/AI Researcher

June 2025 – December 2026

Athens, Georgia

  • Top results across all five homes. Achieved the highest accuracy and weighted F1 on all five smart homes and the highest macro-F1 on four of five (93.2 Aruba, 86.6 Cairo, 84.1 Kyoto, 93.2 Orange4Home; tied on Milan), as measured against 14 published baselines under one shared protocol averaged over five seeds, by splitting each activity episode into four views and training one lightweight expert per view.
  • Gains over the strongest published baseline. Raised macro-F1 over the best published baseline on each home by up to 8.7 points (+8.7 Cairo, +4.1 Kyoto, +2.7 Orange, +1.3 Aruba), as measured on identical stratified 60/20/20 splits, by weighting each expert per episode by how often it was right on similar validation episodes.
  • Efficiency. Beat the strongest multimodal baseline (CARE) by 3.7 to 20.6 macro-F1 points at about one-sixth of its inference cost, as measured by 591 versus 3,631 MFLOPs and 8.2M versus 11.2M parameters, by replacing joint multimodal training with independently trained experts and a parameter-free fusion rule.
  • No trained fusion network needed. Outperformed four jointly trained fusion designs (Concat+MLP, fusion-token transformer, one-stage SupCon, fine-tuned two-stage alignment) by 1.3 to 2.0 pooled macro-F1 points, and a frozen two-stage alignment by 7.3, by combining frozen experts with local competence weighting, which also lets any expert be swapped without retraining.
  • Showed where the gain comes from. Demonstrated that the improvement comes from episode-specific weighting, as measured by a net +121 correct predictions (328 errors fixed versus 207 introduced) on the 26.5% of episodes where experts disagree, and 10.3 percentage points more weight on correct experts, by running leave-one-expert-out and per-episode weight diagnostics.

Stack: Python, PyTorch, Machine Learning, Multimodal Fusion, Sensor Intelligence.

Stylabs AI Native Studio, Full Stack Developer

May 2023 – June 2024

Dubai, UAE

  • Developed more than 50 new frontend components and integrated them with Node.js and Express APIs.
  • Wrote 70+ API calls and MongoDB aggregate queries for data-heavy screens, and created 100% of API documentation used by frontend and mobile developers.
  • Developed the chat support system for customers, employees and owners in the Manzil and Hirevilla rental marketplaces.
  • Tested features, found bugs and fixed them across the client and server.
  • Built 40+ MongoDB aggregation pipelines over 100K+ user and transaction records, reducing average query latency from 750 ms to 350 ms.
  • Shipped 35+ production features across 4 web applications serving 500K+ users.
  • Helped set up the Profound Properties platform: framework choice, template conversion, authentication and data seeding.

Stack: TypeScript, JavaScript, Vue 2/3, React, Next.js, React Native, Tailwind CSS, Node.js, Express, MongoDB, RESTful APIs, Git, Jira, Postman, Swagger.

Lykan Technologies, Full Stack Developer

July 2021 – March 2023

Mumbai, India

  • Reduced average API response times by 35% across high-traffic endpoints by implementing asynchronous request handling in FastAPI and optimizing PostgreSQL query execution plans.
  • Reduced AWS infrastructure costs by 15% by optimizing Docker container image sizes and tuning Kubernetes resource requests/limits.
  • Automated deployment pipelines using Git and Docker, reducing deployment cycle times from 50 to 20 minutes while ensuring zero-downtime releases.
  • Built high-throughput background workers using Python and AWS services to process 10000s records daily with a 99.9% completion rate.
  • Designed a shared React component library with strict TypeScript interfaces, reducing frontend UI bugs by 25% and accelerating feature delivery across teams.

Stack: Python, TypeScript, FastAPI, PostgreSQL, Redis, RESTful APIs, React, Tailwind CSS, Redux, Docker, Kubernetes, AWS, Git, CI/CD, Prometheus, Grafana, Postman.

Freelance, Freelance Developer

January 2020 – June 2021

Remote

  • Built web applications and custom websites for local businesses and organizations.
  • Developed an attendance system for online classes during lockdown to streamline remote attendance tracking.
  • Built a COVID-19 real-time dashboard for tracking regional pandemic data.

Stack: React, JavaScript, HTML, CSS, PHP, WordPress.

Education

  • M.S. Computer Science, University of Georgia

    August 2024 – Expected December 2026

    Athens, Georgia

    • Thesis: Multimodal Self-Supervised Activity Recognition on CASAS Smart-Home Sensor Streams, advised by Dr. Fei Dou.
    • GPA 3.57 / 4.00.
    • Research focus: Ambient sensor intelligence, contrastive representation learning, and edge IoT models.

    Coursework: Algorithms, Distributed Computing, Database Management, Computer Networks, Machine Learning in IoT.

  • B.E. Computer Engineering, Savitribai Phule Pune University

    August 2019 – June 2023

    Pune, India

    • Graduated First Class with Distinction · GPA 8.7 / 10.0.
    • Final Year Project: Resume Ranking System using Machine Learning & NLP (published in IJSREM).
    • Core foundation in data structures, systems architecture, and distributed systems.

    Coursework: Data Structures & Algorithms, Operating Systems, Cloud Computing, Database Systems, High Performance Computing, Machine Learning.

Selected projects

  • Latest project

    HomeSense

    Thesis Platform, June to October 2026

    Smart-home intelligence platform built around my thesis ML model: live Kafka streaming, anomaly detection with human review, LangGraph assistant, digital-twin floorplan viewer, and Kubernetes deployment.

    • Built an end-to-end platform around my thesis model: Kafka streaming, FastAPI, React digital-twin viewer, resident simulator, anomaly review, LangGraph assistant and Kubernetes deployment.
    • Replaced joint multimodal training with independently trained lightweight view experts and local competence weighting, outperforming 4 jointly trained fusion baselines while running at ~1/6th inference cost (591 vs 3,631 MFLOPs).
    • Achieved state-of-the-art results across 5 CASAS smart-home testbeds (+8.7 macro-F1 on Cairo, +4.1 on Kyoto, +2.7 on Orange4Home, +1.3 on Aruba).

    Stack: Python, FastAPI, Kafka, PostgreSQL, React, PyTorch, Kubernetes, Redis, Helm, MLflow.

Work projects

  • Profound Properties

    Production Web Platform, 2024

    Production Dubai real-estate web platform for buying, renting and listing luxury property. Built Next.js frontend components, backend integration with authentication, and AI-assisted data seeding covering 50K records.

    Role: Full-stack contributor. Set up the project foundation, created reusable components from scratch, and implemented AI data seeding that cut manual curation by 40 hours.

  • Manzil

    Live Rental Product, 2023 to 2024

    A live product for design-led holiday homes and service apartments in Dubai. Built real-time multi-role chat support for guests, staff, and owners, along with rental, sales, and expense modules.

    Role: Full-stack contributor. Developed the real-time support chat system for customers, employees and owners, and contributed rental, sales and expense features.

  • Hireavilla

    Live Rental Marketplace, 2023 to 2024

    Luxury vacation rental marketplace in India. Developed unified multi-role support chat system for guests, property owners, and operations staff, plus core rental and expense modules.

    Role: Full-stack contributor. Developed support chat and core modules across both customer-facing applications and operations portals.

  • The Mainstreet Marketplace

    Live Commercial Product, 2023 to 2024

    High-scale sneaker and streetwear e-commerce platform. Developed frontend components, API contracts, and performance optimizations cutting page load times from 4s to 1s.

    Role: Full-stack contributor. Developed components and API integrations, fixed bugs, and authored comprehensive API documentation improving user engagement by 27%.

Research & Publications

Skills

Languages
Python, TypeScript, JavaScript, SQL, C, C++, Java, PHP, HTML/CSS
Machine learning
PyTorch, PyTorch Geometric, Multimodal Fusion, Self-Supervised & Contrastive Learning, Transformers, Graph Neural Networks (GNNs), CNNs, scikit-learn, NumPy, pandas, matplotlib, MLflow, LangChain, LangGraph, RAG
Backend and data
FastAPI, Node.js, Express, Flask, RESTful APIs, Kafka, PostgreSQL, MongoDB, Redis, MySQL, Socket.IO, Drizzle ORM
Cloud and DevOps
Docker, Kubernetes (EKS), Helm, AWS (S3, EC2, RDS), GitHub Actions (CI/CD), Prometheus, Grafana, Linux / Bash
Frontend
React, Next.js, React Native (Expo), Vue 2/3, Nuxt.js, Vite, Tailwind CSS, Redux, KaTeX
Systems and networking
TCP & UDP sockets, Threads & Locks, Distributed KV stores, Consistent hashing, Discrete-event simulation, DNS & HTTP range requests, OpenSSL
Testing and tools
Git, Postman, Swagger, Vitest, Playwright, Jira, Linux

Certifications

  • Google Cloud

    • Manage Kubernetes in Google Cloud
    • Create and Manage Cloud SQL for PostgreSQL Instances
  • AMD AI Academy

    • Kubernetes on AMD Instinct GPUs - From Zero to AI Inference
  • DataCamp

    • AI Engineer for Developers Associate
  • GitHub

    • GitHub Foundations
  • Databricks Academy

    • Databricks Accredited Generative AI Fundamentals
  • IBM SkillsBuild

    • Web Development Fundamentals
    • Data Fundamentals
    • Artificial Intelligence Fundamentals
  • Claude Academy

    • AI Fluency for Builders
    • Building with the Claude API
    • Claude 101
    • Claude Code 101
    • Introduction to Claude Cowork
  • OpenAI Academy

    • AI Foundations
  • EY and Microsoft

    • AI Skills Passport

Additional Systems & Course Projects

  • myDocker

    Developer Tool, May 2026

    Local Docker control plane that automatically discovers compose projects, allocates non-colliding host ports, generates overrides without touching repo files, and streams container logs and builds live in the browser.

    Role: Designed and built the dashboard, non-colliding port allocator, streaming terminal, and Compose execution layer.

  • Ringside Watch-Party

    Real-time Platform, August 2026

    Real-time synchronized sports watch-party platform with spoiler-safe delayed chat, reaction counters, live UFC/F1/cricket rankings, and cross-platform Android mobile app. Used by ~200 users.

    Role: Built the web application, real-time Socket.IO server with reaction batching, normalized sports API adapters, and Expo mobile app.

  • MLForge

    Interactive Learning Tool, September 2026

    Browser-only machine learning workbench with 75 interactive labs and 20 guided projects. Visualizes neural network training, decision boundaries, and loss curves in real-time with from-scratch TypeScript algorithms. Used by 5,000+ users.

    Role: Designed and implemented the interactive lesson system, all from-scratch ML algorithms in TypeScript, and the Vite client.

  • SystemForge

    Interactive Learning Tool, September 2026

    Interactive distributed systems learning lab and simulator featuring 40 system labs, an architecture canvas, failure injection, Raft consensus, and event streaming playgrounds. Helped over 10,000 learners.

    Role: Designed and built the simulation engine, 40 system labs, canvas, code playground, and Next.js web application.

  • Federated Face Recognition Robustness

    Graduate Research Project, 2025

    Study of adversarial attack vulnerabilities in smart-home face recognition trained under federated learning. Evaluated FGSM and Square attacks on the FLATS dataset and proposed a hybrid defense combining denoising autoencoders with reverse-attack strategies.

    Role: Team lead. Coordinated experimental design, adversarial perturbation testing, and evaluation across four team members.

  • Java relational algebra engine executing select, project, union, minus, and join operators. Optimized with LinHashMap indexing and predicate selectivity-aware query planning.

    Role: Implemented relational operators from scratch, integrated linear hash index structures, and added predicate-aware planning.

  • Distributed Key-Value Naming Service

    Distributed Systems, 2024

    Java consistent hashing hash-ring service with a bootstrap node and dynamic name servers. Handles dynamic node joins/leaves by reassigning affected key ranges on a 1024-position ring over TCP sockets.

    Role: Wrote the bootstrap server, dynamic name server range reassignment, and TCP socket command client individually.

  • Resume Ranking with ML and NLP

    Undergraduate Research, January to June 2023

    End-to-end resume ranking platform using machine learning and NLP algorithms connected to a Flask API and React interface. Published as a technical paper in IJSRMS (May 2023).

    Role: Team lead. Built the Flask server, React client, and ML model integration, and co-authored the technical paper.