cv

BS

Indian Institute of Technology, Jodhpur

Applied AI and Data Science

• Core focus on Deep Learning, Linear Algebra, and Statistical Inference.

MicroMasters

MITx

Statistics and Data Science

• Advanced training in Probabilistic Modeling, Machine Learning foundations, and Time Series Analysis.

LLM Lab

• Built a character-level GPT from scratch in PyTorch to study the internal dynamics of Multi-Head Attention and Transformer blocks.
• Implemented residual connections and layer normalization, stabilizing gradient flow and training convergence.

Lenia & Conway's Game of Life

• Simulated continuous-state artificial life via Gaussian kernel evolution and non-linear PDEs, probing emergent dynamics at real-time high resolution.

rCiv (Knowledge Commons)

• Built a verified knowledge commons for AI-generated research, a FastAPI service where agents submit source-referenced claims that resolve into a canonical temporal knowledge graph.
• Engineered source-aware LLM deduplication and conflict detection over Graphiti + Neo4j, fusing semantic, BM25, and graph search served via RAG.

ARC-Engine

• Developed a high-performance OpenGL rendering engine with a custom Entity Component System (ECS) for cache-friendly memory layout.
• Built a real-time visualization pipeline with custom compute kernels for low-level, compute-heavy workloads.

Tailor (AI Framework)

• Built a modular AI framework in Rust (Tauri) and Python that lets users stitch models and tools into custom assistants.

Sudoku Solver (Algorithms)

• Investigated Backtracking vs. Bitmasking approaches for combinatorial optimization, pruning the search space with heuristics to cut runtime by 80%.

AI Engineer Intern

Hapticware Intelligence

• Built an agentic swarm framework with LangGraph to orchestrate and manage a fleet of 30+ agents across complex, decomposable tasks.
• Engineered RAG over a 100+ table production database, cutting chatbot query response time by 30-40% and auto-generating result graphs.
• Developed backend APIs for a digital-signer product and its admin portal, alongside supporting production services.

Research Intern

Abhay Travels

• Researched demand-forecasting models for time-series data, engineering 15+ features to capture seasonal and non-linear trends.
• Built a predictive pipeline with CatBoost that beat baseline ARIMA models on RMSE across 5 years of historical booking data.

Data Analyst Intern

AMR India

• Forecast article-specific demand cycles across 200+ HSN codes from ~50K port transaction records over 4 years, informing quarterly procurement planning.