Bengaluru, India

Ishaan Snehal
Parikh

> _

Dual-degree undergrad (CS @ RVCE, Data Science @ IIT Madras) building agentic AI systems — from production KYC platforms to AI-aware OS schedulers.

View Projects Download Resume

01 / About

Two degrees, one obsession: building.

I'm a dual-degree undergraduate reading B.E. Computer Science & Engineering at RV College of Engineering (CGPA 9.62) alongside a B.S. in Data Science & Applications at IIT Madras (CGPA 9.00) — running both in parallel from Bengaluru.

My foundation is in DSA, Operating Systems, Computer Networks, and DBMS, and I spend most of my time shipping across agentic AI, machine learning, systems programming, computer vision, and full-stack AI development. I like problems that sit where research meets production.

On the side, I co-run Gujarat Junction, a cloud kitchen — which keeps me honest about margins, operations, and shipping things people actually pay for.

RVCE — CS&E

9.62

CGPA

IIT Madras — DS

9.00

CGPA

Based in

Bengaluru

Also building

Gujarat Junction

cloud kitchen

02 / Experience

Where I've been building.

Feb 2026 – Present

Product Development Intern — “Drawganizer”

BerkeleySoft USA

  • Engineered an AI drawer-organization system using credit-card homography calibration (ISO/IEC 7810) to convert image pixels to real-world millimeters without depth sensors or LiDAR — 88% L/W and 97% height calibration confidence.
  • Built a Union-Find-based compartment detection engine to auto-segment drawer layouts, paired with a 3D preview and a CNC G-code exporter for cross-lap joinery.

Nov 2025 – Jun 2026

AI Research Intern — AIOS Scheduler Research

Samsung R&D Institute India (SRI-B) · Samsung PRISM

  • Designed and benchmarked 7 scheduling algorithms (FCFS, SJF, SRTF, Round Robin, Lottery, Priority, MLFQ) for concurrent on-device AI workloads (GPT-2, BERT-Base, OPT-125M) on the ONNXim NPU simulator across 89+ interference tasks.
  • Lifted deadline satisfaction from ~1% (FCFS) to 100% (Preemptive Priority); MLFQ reached 94.4% deadline compliance while cutting lightweight inference latency from >900ms to ~4ms.

Aug 2025 – Oct 2025

Agentic AI Engineer Intern

Tata AIA Life Insurance

  • Built a production-ready FastAPI + LangGraph multi-agent e-KYC chatbot spanning 5 document types (PAN, Aadhaar, Passport, DL, Form 60) with OCR and RAG-based classification — cutting manual verification 60–70% and onboarding time 40–50%.
  • Designed an L1/L2/L3 hierarchical memory architecture (Redis + vector DB) for session-persistent conversations, reducing KYC decision time by 35–40%.

2024 – Present

Co-founder & COO — Cloud Kitchen

Gujarat Junction

  • Runs P&L, vendor sourcing, menu pricing, and Zomato/Swiggy operations end to end.
  • Built margin-tracking frameworks to keep the business profitable while scaling.

03 / Projects

Things I've shipped.

FraudGraph

Real-time financial fraud-ring detection.

Live graph pipeline using Tarjan's SCC + PageRank to flag fraud rings from streaming transactions (9-node rings, 6.3% fraud ratio). Differentially-private Random Forest with SHAP/LIME explainability, evaluated on 1M+ transactions.

PythonXGBoostWebSocketGraph ML

AI Land Cover Change Detection

Western Ghats deforestation monitoring.

U-Net (ResNet-34 encoder) trained on DeepGlobe reaching 0.714 mIoU / 87.86% pixel accuracy, with live Sentinel-2 integration. React dashboard renders year-on-year transition matrices and auto-generates PDF conservation reports.

PyTorchFastAPIGoogle Earth EngineReact

Decentralised Multimodal Biometric Auth

Identity without a local database.

Fully decentralized identity fusing ArcFace (face), ECAPA-TDNN (voice), and OCR (document) embeddings, AES-256-CBC encrypted on IPFS. Ethereum smart contracts map DIDs to IPFS CIDs, compressing 4MB biometric payloads to a 32-byte on-chain hash.

PythonIPFSEthereum SepoliaCryptography

More on GitHub

github.com/ishaanparikh14

04 / Skills

The toolkit.

Languages

PythonCC++Embedded CBash

AI / ML

Agentic AIComputer VisionNLPRandom ForestXGBoostLSTMSemantic SegmentationData Analysis

Systems

Linux/UnixTCP/IPNetwork SecurityOperating SystemsGit

Frameworks

FastAPIFlaskReactLangGraphRedisPyTorchONNX Runtime

Core CS

DSADBMSComputer NetworksOOP

05 / Recognition

Achievements & education.

Achievements & Leadership

Head of Operations, ACM RVCE Student Chapter
Senior Associate, Entrepreneurship Cell RVCE
Student Member, IEEE Computer Society RVCE — MobileViT-Graph satellite road extraction (IAmPro 2026)
Open Source & Dev Team, Google Developer Groups RVCE
99% Class X · 97% Class XII
JEE Mains — 98.5 percentile
COMEDK AIR 218 · KCET AIR 899
Certifications: CCNA · NPTEL Elite+Silver (Design, Tech & Innovation) · SEBI Investor · Microsoft Power BI Foundational

Education

IIT Madras

CGPA 9.00

B.S. Data Science & Applications

Jan 2025 – Dec 2028

RV College of Engineering

CGPA 9.62

B.E. Computer Science & Engineering

Sep 2024 – Jun 2028

06 / Contact

Let's build something.

Working on agentic AI, systems, or ML and want an extra pair of hands? Have a role, a project, or just a good problem? My inbox is open.

ishaan.isp11@gmail.com