HOME
PROFILE //

Paras Goel

I'm a computer science and applied mathematics student at Columbia University interested in building intelligent systems at the intersection of AI agents, generative models, and systems engineering. Currently working on Copilot Frontier Tuning at Microsoft and video generation at DitecT Lab.

TECHNICAL // STACK & SKILLS

What I Work With

LANGUAGES
Python C++ Java C TypeScript JavaScript SQL Bash MATLAB HTML/CSS
AI & MACHINE LEARNING
PyTorch Hugging Face (Diffusers, Accelerate) CUDA LangGraph PydanticAI LangChain MONAI Vision Transformers Diffusion Models Scikit-learn TensorFlow / Keras Weights & Biases
SYSTEMS & CLOUD
Model Context Protocol (MCP) Docker Azure (AI Foundry, ADX) Google Cloud (GCP) Redis Qdrant (Vector DB) Ray Slurm PostgreSQL Supabase Grafana FFmpeg
WEB & FRAMEWORKS
React Next.js Astro FastAPI Node.js REST APIs OAuth Vercel Git Tailwind CSS Postman Flask

Software Development Intern, Copilot Tuning Team

Microsoft Redmond, WA
May 2026 – Present
  • Developed evaluation and experimentation infrastructure for the Copilot Frontier Tuning (FT) platform, enabling reinforcement learning based optimization of enterprise AI agents and LLMs.
  • Built & scaled performance analytics for 1,000+ agents across 250+ FT tenants, used daily by 30+ developers.
  • Leveraging Monte Carlo Tree Search to evaluate FT models' reasoning, latency, and token efficiency tradeoffs.
  • Developing an autonomous AI agent to orchestrate feature rollout workflows across 33,000+ flights, including tenant onboarding, deployment debugging, & configuration validation.
Azure Data Explorer AI Foundry RL World O365 Monitoring WorkIQ Graph API

ML Researcher

DitecT Lab, Columbia University Engineering New York, NY
Jan 2026 – Present
  • Developing causal inference frameworks for video generation by modeling temporal dependencies and intervention effects, enabling more controllable and interpretable generation dynamics.
  • Adapting causal forcing techniques to distill large bidirectional generative models into efficient autoregressive architectures, reducing inference cost while preserving generation quality.
  • Analyzing temporal consistency, causal fidelity, and generation efficiency across video synthesis models.
CUDA Hugging Face Diffusers Accelerate Ray Slurm FFmpeg Hydra NumPy

AI Engineer

Agentic Fabriq (YC F25) New York, NY
Jan 2026 – May 2026
  • Developed OAuth-based authorization workflow for enterprise AI agents, enabling secure access to customer resources while maintaining isolation across agent workflows.
  • Designed and deployed MCP servers and tool-integration layers supporting authenticated communication between LLM agents, external services, and internal data systems.
MCP FastAPI Redis Qdrant Google Cloud LangGraph PydanticAI Weights & Biases

Software Developer

PayPal San Jose, CA & New York, NY
Jun 2025 – Dec 2025
  • Leveraged LLMs to transform unstructured business content into production-ready storefronts, enabling rapid merchant onboarding with generated catalogs & integrated payments (Alpha tested with 100+ customers).
  • Developed infrastructure for full-stack system by integrating multimodal image generation models, LLMs, and PayPal SDKs with responsive UIs and scalable cloud deployment on Vercel.
  • Created evaluation pipelines for AI models' outputs for catalog creation, pricing, natural language-to-HTML conversion, and price accuracy.
Next.js Node.js Vercel PostgreSQL React LangChain OAuth REST APIs

AI/ML Researcher

Columbia University Irving Medical Center New York, NY
May 2024 – Nov 2025
  • Implemented 3D attention mechanisms for glioma classification from multimodal MRI scans.
  • Built image-to-image deep learning pipeline using latent diffusion model for Region of Interest extraction.
PyTorch Scikit-learn TensorFlow/Keras Docker Vision Transformer Pydicom Grafana
NeurIPS 2025

Utilizing MCP and Shared Introspection for Targeted Agent-to-Agent Communication in Hierarchical Clinical Multi-Agent Systems

We developed a hierarchical multi-agent system for clinical decision support, combining five domain-specialist agents with an attending-style meta-agent to coordinate complex medical reasoning. Through structured deliberation and confidence-based conflict resolution, our system improved diagnostic accuracy from 20% to 80% on USMLE-style clinical questions while reducing response time by approximately 28%.

IEEE RAAI (Robotics, Automation, and Artificial Intelligence) 2025

Facilitating Shared Introspection in Hierarchical Multi-Agent Systems Using the Model Context Protocol

We developed an MCP-enabled multi-agent framework that allows specialized medical AI agents to share and reflect on their reasoning in real time. By coordinating these insights through a supervising agent, we explored how structured introspection and deliberation can make clinical AI more transparent, efficient, and resilient to individual agent errors.

AJAS 2025

Utilizing Multi-Agent Reinforcement Learning with Encoder-Decoder Architecture Agents to Identify Optimal Resection Location in Glioblastoma Multiforme Patients

We developed an end-to-end AI framework for glioblastoma diagnosis and treatment planning, combining sequential classification with generative models that simulate surgical resection, radiotherapy, and chemotherapy outcomes. By integrating these models with reinforcement learning, our system iteratively explores treatment strategies toward a desired survival outcome while substantially reducing diagnostic compute costs and tumor-progression inference time.

IEEE ISBI (International Symposium on Biomedical Imaging) 2024

CoCa-Mil: Attention-Based Handcrafted-Deep Feature Fusion in Computational Pathology

We developed CoCa-MIL, an attention-based framework for whole-slide image classification that combines domain-specific handcrafted features with representations learned by deep neural networks. Using co-attention and cross-attention mechanisms, our approach leverages the complementary strengths of both feature types, improving classification accuracy on the TCGA Lung Cancer dataset by up to 5.21% over baseline methods.

IEEE SPIE (Society of Photographic Instrumentation Engineers) 2024

CA-fuse-MIL: cross-attention fusion of handcrafted and deep features for whole slide image classification

We introduced CA-Fuse-MIL, a cross-attention approach that integrates handcrafted pathology features with deep learned representations for whole-slide image classification. We also explored multi-layer variants of the architecture, showing that explicitly learning interactions between these complementary feature spaces can outperform deep features alone, with accuracy gains of up to 5.21% on the TCGA Lung Cancer dataset.

IEEE SPIE (Society of Photographic Instrumentation Engineers) 2023

Role of stain normalization in computational pathology: use case in metastatic tissue classification

We investigated how stain normalization and color augmentation can improve the robustness of deep learning models across variations in histopathology imaging. Evaluating six preprocessing configurations on more than 300,000 tissue images, we found that combining Macenko normalization with color augmentation produced the strongest metastatic tissue classification performance, improving both accuracy and F1 score over the baseline.

BACKGROUND // EDUCATION & LEADERSHIP

Education & Involvements

EDUCATION

Columbia University

Fu Foundation School of Engineering and Applied Science
B.S. in Computer Science & Applied Mathematics
Minor in Statistics • 3.81 GPA • Dean's List (Expected 2028)
SELECTED COURSEWORK
Machine Learning Artificial Intelligence Natural Language Processing Computer Vision Analysis of Algorithms Data Structures Advanced Programming Operating Systems Databases Security CS Theory Discrete Mathematics
ORGANIZATIONS & ACTIVITIES
Columbia Space Initiative CubeSat Team
CU Airplane Club Electrical Subteam
Engineering Recruitment Council Undergraduate Tour Guide
CU Science Olympiad Event Supervisor
CURRENT FOCUS

Generative models, multi-agent reasoning, causal inference, and building production-scale AI architectures.