About me

AI engineer building agentic systems, data pipelines and LLM-based automation that run in production. M.S. in Computer Science from Rutgers University (GPA: 3.65).

At Beacon Health AI, I build and run agentic workflows in production: a fax-processing pipeline that has handled ~9,000 clinical documents since May 2026, and an EHR workflow across Athena, Medent and Elation for 3 primary care centers under HIPAA. On my own, I built LeanJobOS, a full-stack AI job-search platform with an LLM router that keeps AI cost to about $0.07 per application.

Before that, I built generative-AI tools on AWS Bedrock and QuickSight dashboards for the New Jersey Department of Labor, and ETL pipelines over 8M+ SEC EDGAR filings on Rutgers' Amarel HPC cluster. At Code-blue AI (UC Berkeley–affiliated), I built real-time stroke detection with sensitivity and specificity above 90% and raised a speech-recognition model to 98.9% accuracy. My capstone, Core Sentinel, runs a quantized TinyBERT model on-device in under 100 ms end to end.

On the analytics side, I grew website traffic 3× in one quarter at the Rutgers–Camden School of Business and led a GitHub Copilot rollout that delivered a 20% gain in delivery velocity for a corporate client.

Stack: Python, SQL, C++, AWS, GCP, Docker, ONNX, LLMs, NLP, LangChain, RAG pipelines, ETL, Power BI, Shell Scripting

Published researcher — IEEE ICRITO 2024 (Best Paper Award). President, Rutgers Graduate Student Organization. AWS Solutions Architect Associate (Expected September 2026). Actively targeting full-time roles in AI Engineering, ML Infrastructure, and Data Engineering — willing to relocate.

What I'm Doing

  • Production AI & ML

    Real-time inference, ONNX-quantized models, LLM and RAG pipelines, LangChain, and shipping multimodal systems to production with measurable accuracy and latency targets.

  • Compliance & Guardrails

    Enterprise agents and desktop tooling aligned with HIPAA, FERPA, and SEC-style policy: local inference, zero exfiltration, and sub-100ms guardrails where the workload demands it.

  • Data Pipelines & Cloud

    Large-scale ETL on AWS and GCP, Dockerized services, HPC-oriented batch jobs, and automated NLP over large archives (e.g., 8M+ SEC regulatory filings).

  • Analytics & Impact

    GA4 and marketing analytics at scale, executive-facing reporting with Power BI, and measurable lifts in efficiency and acquisition through experiment-driven optimization.

Community & Leadership

  • Student Community Lead @ Progate

    Developed and managed the overall student developer community. Organized campus hackathons, conducted technical workshops on web programming, and expanded educational reach through self-initiated and pre-existing company initiatives.

  • Campus Ambassador @ TechLearn

    Promoted technical educational courses across campus networks. Drove enrollment growth through digital marketing strategies and leveraged data visualization to analyze and optimize campaign reach and conversion rates.

  • President @ Rutgers Graduate Student Organization

    Elected President managing advocacy, student governance, and campus events. Coordinated outreach to 1,500+ incoming graduate students, raising awareness for campus resources and support programs. Organised the first-ever RU Hack on campus co-hosted with Major League Hacking (MLH), drawing 60+ participants.

  • Cohort Leader @ Rutgers AI Campus

    Led development in one of the first three AI cohorts across campus. Directed technical support for the Rutgers Archaeology Department on the Arch Street Project, streamlining excavation data, constructing interactive 3D spatial mapping clusters, and assisting archaeologists in analyzing findings.

Developer Badges

Resume

Publications & Research

Google Scholar Profile
  1. Quantum Networking: Creating the Future Landscape of Refined Classical Communication

    🏆 Best Paper Award
    11th IEEE International Conference on Reliability, Infocom Technologies and Optimization (ICRITO 2024)
    March 2024 • Cited by 5
    • • Authors: Ayush Kumar Sharan, Alwin Philip, Purushottam Sharma, Vinod Kumar Shukla
    • • Explored integration of Quantum Key Distribution (QKD) and quantum network topologies with classical communication systems to refine network reliability and defend against quantum cryptographic threats.
    • • View on Google Scholar
  2. Avatars in the virtual realm: An integrated approach to enhance digital community interaction

    International Journal of System Assurance Engineering and Management (Springer)
    2024 • DOI: 10.1007/s13198-024-02560-z • Cited by 4
    • • Authors: Punj Lata Singh, Nitasha Hasteer, Rahul Sindhwani, Alwin Philip, Rishika Sharma
    • • Proposed an integrated framework for virtual avatars in metaverse environments across tourism, healthcare, research, and social networking while addressing privacy and cybersecurity challenges.
    • • View Springer Article | View on Google Scholar

Education

  1. Rutgers University

    2024 — 2026
    • Master of Science in Computer Science
    • GPA: 3.65
    • Relevant Coursework: Big Data Algorithms, Data Structures, Machine Learning, Distributed Systems, Cryptography, Software engineering
  2. Amity University

    2020 — 2024
    • Bachelor of Technology in Information Technology
    • GPA: 3.82

Experience

  1. Agentic AI Engineer (Contract)

    Beacon Health AI, Brooklyn, NY
    May 2026 — Present
    • • Create new agentic workflows and debug and improve the agentic harness that runs them in production.
    • • Own a production fax-processing pipeline built around an automated fax-filing agent, handling 450–500 clinical documents a week (~9,000 since May 2026).
    • • Own an EHR data workflow integrating Athena, Medent and Elation that automates patient scheduling and follow-ups for 3 primary care centers under HIPAA.
    • • Audited agent run-history logs against Cyberdesk (source of truth), resolving data discrepancies with the engineering team.
  2. Software Development Specialist (Analytics/AI)

    New Jersey Department of Labor and Workforce Development, Trenton, NJ
    June 2026 — August 2026
    • • Built interactive AWS QuickSight dashboards on the department's AWS data lake, giving leadership and program staff self-serve views of workforce and labor-market trends.
    • • Built dashboards on Salesforce-integrated data to analyze unemployment insurance trends across New Jersey.
    • • Built generative-AI applications on AWS Bedrock, including prompt-engineering workflows used across department operations.
  3. Graduate Research Assistant - Systems and Data Engineering

    Rutgers University-Camden, Camden, NJ
    Dec 2025 — May 2026
    • • Architected an automated Python and Shell ETL pipeline over 8M+ SEC EDGAR filings (10-K, 8-K, 6-K), structuring unstructured financial text for downstream auditing models and screening for environmental covenants.
    • • Engineered preprocessing automation as scheduled, repeatable batch jobs on Rutgers' Amarel HPC cluster, handling large financial and behavioral datasets with minimal manual intervention.
    • • Applied NLP techniques to extract structured signals from legacy financial filings, improving model input quality and auditability for research stakeholders.
  4. Data Science & AI Research Intern

    Code-blue AI, UC Berkeley-affiliated Startup
    June 2025 — August 2025
    • • Built real-time stroke detection from facial-asymmetry and speech-pattern analysis, with sensitivity and specificity above 90%.
    • • Raised the speech-recognition model to 98.9% accuracy by expanding and diversifying its training data to nearly 17,000 audio samples from the TORGO and UA-Speech datasets.
    • • Built the high-volume facial and speech stream ingestion and real-time inference pipeline for stroke detection.
    • • Technologies used: Python, TensorFlow, PyTorch, Computer Vision, NLP
  5. AI Training & Prompt Engineering Specialist

    Rutgers Business School & Thayer Distribution
    June 2025 — July 2025
    • • Led the GitHub Copilot and Microsoft Copilot rollout at Thayer Distribution, training employees to automate workflows and measuring a 20% gain in delivery velocity.
    • • Designed and delivered prompt-engineering and AI-tool training for professional and corporate audiences.
    • • Created training materials and best practices documentation for enterprise AI adoption.
    • • Technologies used: GitHub Copilot, Microsoft Copilot, AI Prompt Engineering, Workflow Automation
  6. Marketing Analytics and AI Specialist

    Rutgers-Camden School of Business
    August 2024 — May 2026
    • • Grew website traffic 3x in one quarter while managing a $40K marketing budget.
    • • Built and maintained GA4 and Looker Studio dashboards used daily by leadership, run as always-on production reporting.
    • • Maintained GA4 tracking and web forms, and analyzed traffic sources, engagement and user behavior to improve conversion funnels and turn inquiries into leads.
    • • Ran campaign performance analysis and audience segmentation across web, email and social channels, and worked on SEO and UI/UX improvements to the WordPress site.
  7. Mobile App Developer Intern

    Aeologic Technologies, Noida
    May 2023 — July 2023
    • • Designed and built UI for a shopping app in Flutter and Android Studio, including a checkout flow that ran 15% faster.
    • • Improved app responsiveness by 25% through code reviews and testing with the development team.

Skills

  • AI & LLMs

    Claude API, AWS Bedrock, agentic workflows, RAG, LangChain, prompt engineering, PyTorch, TensorFlow, scikit-learn, ONNX Runtime, OpenAI Whisper

  • Data Engineering

    Python, SQL, Shell, ETL pipelines, API ingestion, scheduled batch jobs, Amarel HPC, PostgreSQL, MySQL, SQLite, MongoDB

  • Cloud & DevOps

    AWS (S3, Lambda, EC2, Athena, QuickSight, Bedrock), GCP, Oracle Cloud, Docker, Git, CI/CD, Linux

  • Full-Stack

    FastAPI, Django, React, TypeScript, Node.js, REST APIs, Playwright

  • Analytics & BI

    AWS QuickSight, Looker Studio, GA4, Power BI, Excel, statistics

Portfolio

Blog

Contact

Contact Form