# Hari Om Dwivedi

> ML & AI Engineer (SDE) building production LLM, RAG, semantic search and Apache Spark systems. Cut analytics turnaround from 48 hours to under 1 minute. Open to Machine Learning Engineer, AI Engineer, LLM Engineer, Software Development Engineer (SDE) roles.

- Website: https://www.hariom.work/
- Email: dwivedihari987@gmail.com
- LinkedIn: https://www.linkedin.com/in/HariomDwivedi
- GitHub: https://github.com/HariOm987
- Résumé (PDF): https://www.hariom.work//Hari_Om_Dwivedi_Resume.pdf
- Location: India

## Highlights
- 48h → <1 min — Analytics turnaround (Conversational BI · Carnation Infotech)
- +25% — Lead-to-client conversion (Embedding-based ranking · Propques)
- $10,000 — Shopify bounty (Two production Shop Minis)
- 700+ — Students mentored (President, CSI Student Chapter)

## Experience

### Software Development Engineer I — Carnation Infotech (Oct 2025 — Present)
Conversational AI for Business Intelligence
- Designed and built an enterprise-scale conversational AI platform for the Business Intelligence team, enabling natural-language interaction with company-wide big data.
- Architected an LLM-powered query translation layer that converts stakeholder questions into optimized PySpark SQL queries.
- Integrated secure query execution on Apache Spark clusters with automated aggregation, validation and post-processing of results.
- Developed summarization and visualization pipelines that return charts, tables and insights in a user-friendly format.
- Implemented caching, query optimization and guardrails to ensure performance, correctness and cost efficiency.
- Reduced analytics request turnaround from 48 hours to under 1 minute, dramatically improving stakeholder productivity.
- Collaborated with data engineering, platform and leadership teams to enforce access control, data governance and scalability.

### Software Development Engineer Intern — Carnation Infotech (Aug 2025 — Oct 2025)
Karmik — AI recruitment & talent intelligence
- Built a robust résumé-parsing engine that extracts structured data from unstructured PDF and DOCX résumés.
- Implemented semantic and keyword-based advanced search for efficient candidate discovery at scale.
- Designed and deployed a candidate recommendation system using embeddings and ranking algorithms.
- Led migration of the core database from MongoDB to PostgreSQL, including schema redesign and data backfill.
- Optimized queries and indexing strategies to improve search latency and reporting capabilities.

### Software Engineering Intern — Propques (Jul 2024 — Oct 2024)
PropSaaS — real-estate analytics & lead management
- Developed AI-powered lead-intelligence systems using embeddings, similarity scoring and ranking models.
- Designed feature-engineering pipelines that significantly improved lead quality and relevance.
- Improved lead-to-client conversion rates by 25% through data-driven ranking and scoring logic.
- Led backend development of PropSaaS, a real-estate analytics and lead-management SaaS platform.
- Built automated onboarding workflows, CRM integrations and performance-analytics dashboards.
- Containerized backend services and ML inference pipelines with Docker for consistent deployments.
- Worked closely with product, sales and leadership teams to iterate rapidly on features aligned with business KPIs.

## Projects

### Shopify Shop Minis — Help Me Decor & Time Traveller
Two production shopping experiences that live inside Shopify’s Shop app — built in partnership with Shopify and awarded a $10,000 bounty.
- Built and deployed two production-grade Shopify Shop Minis in partnership with Shopify.
- Designed interactive, discovery-driven shopping experiences embedded directly within the Shop app.
- Implemented recommendation logic, UI state management and backend integrations using Shopify Mini APIs.
- Focused on performance optimization, UX polish and scalability for real-world usage.
- Awarded a $10,000 USD cash bounty by Shopify for innovation, impact and execution quality.
- Help Me Decor: https://shop.app/mini/help-me-decor
- Time Traveller: https://shop.app/mini/time-travel

### Whisper — Local GenAI assistant on WhatsApp
A privacy-first assistant that answers questions over WhatsApp with retrieval-augmented generation — the LLM runs locally, not behind a third-party API.
- Built a privacy-first local LLM assistant providing real-time question answering via WhatsApp.
- Implemented retrieval-augmented generation using embeddings and a vector database (ChromaDB).
- Designed ingestion, chunking, retrieval and prompt-optimization pipelines for accuracy and low latency.
- GitHub: https://github.com/Shanvithegreat0/Localbuddy

### Fridge Talks — Smart inventory & meal planner
A computer-vision system that tracks what’s in the fridge, predicts spoilage and nudges you on WhatsApp before food goes bad.
- Built an end-to-end AI system for refrigerator inventory tracking and food-spoilage prediction.
- Implemented computer-vision pipelines using TensorFlow and Mask R-CNN for item detection.
- Designed backend services for expiry modeling, recommendation logic and inventory state management.
- Integrated real-time WhatsApp alerts for reminders, spoilage warnings and meal suggestions.
- GitHub: https://github.com/HariOm987/smart-fridge/

### No Mic — Real-time audience interaction
Live audience voting and Q&A for events, built for low-latency updates with WebRTC and Socket.IO.
- Developed a real-time platform enabling live audience voting and Q&A during events.
- Implemented low-latency communication using WebRTC and Socket.IO with a scalable Django backend.
- GitHub: https://github.com/HariOm987/no-mic

## Skills
- LLMs & Retrieval: LLMs, RAG, Text-to-SQL, Embeddings, Semantic search, Vector DBs (ChromaDB), Prompt engineering, LLM guardrails, Hybrid search, Chunking & ingestion, LLM evaluation, Hugging Face Transformers, LangChain
- Data & Distributed: Apache Spark, PySpark, Spark SQL, ETL pipelines, Data warehousing, Data visualization, Query optimization
- Databases: PostgreSQL, MongoDB, Redis, Cassandra, SQL, Schema design & migration, Indexing, SQLite, Firebase
- ML & Computer Vision: Recommender systems, Ranking & scoring, Feature engineering, Computer vision, Mask R-CNN, TensorFlow, Document parsing, Predictive modeling, PyTorch, scikit-learn, pandas & NumPy
- Backend & APIs: Python, Django, Flask, REST APIs, Microservices, WebRTC, Socket.IO, Caching, Node.js, GraphQL, Testing (unit, integration, API), FastAPI
- Cloud & MLOps: Docker, Kubernetes, AWS, GCP, Azure, CI/CD (Jenkins), Model serving, MLOps fundamentals, Linux, Git
- Languages & Frontend: Java, Kotlin, JavaScript, TypeScript, Bash, React

## Education
- B.Tech, Computer Science & Engineering (First Division), SRMCEM (affiliated with AKTU), Lucknow, 2021 — 2025
- President, CSI Student Chapter — led technical initiatives, organized large-scale hackathons and mentored 700+ students.
- Recognized for leadership, system design and innovation through multiple national-level hackathon awards.

## Awards
- First Prize — Hack-To-Crack, AKTU, Lucknow
- Second Prize — UP Police Hackathon, 2022
- Second Prize — HackOFiesta, IIIT Lucknow
- Most Innovative Hack — HackCBS, Delhi
- Cash bounty — Shopify Shop Minis, Shopify

## Certifications
- Cybersecurity Fundamentals (IBM): https://drive.google.com/file/d/15Uu1AoDOlLx4--XBpZnSZ3tPqbpbINIo/view?usp=sharing
- Data Privacy Fundamentals (Cognitive Class): https://drive.google.com/file/d/1R74MG_kvamW1qsjTevYDLhJ53o5eZTD7/view?usp=sharing
- Introduction to Cloud (IBM SkillsBuild): https://drive.google.com/file/d/1o6Bac8KVuAZ5hIEq_UsguRYyR2nTSAlN/view?usp=sharing
- Data Science for Business — Level 1 (IBM): https://drive.google.com/file/d/1ImrdnSUebkPyGjHW9NeZveqqO91qUY88/view?usp=sharing