Curriculum Vitae

A.1 Education ¶

institution

Indian Institute of Science Education and Research, Bhopal

degree

B.S., Computer Science and Engineering

minor

Data Science and Engineering

period

Dec 2022 – Jun 2026

conferred

25 June 2026

gpa

8.07 / 10.0

A.2 Experience ¶

AI Engineer, iGPT Agency, Bengaluru

period

Jan 2026 – Present

• Engineered an AI filmmaking platform orchestrating 12+ LLM agents across 4 generative-AI APIs, cutting per-scene production time 70–85% by replacing a manual prompt-and-regenerate workflow with a single gated pipeline featuring reference-locked continuity and one-click assembly.
• Built an end-to-end candidate-intelligence platform that scrapes LinkedIn profiles via Playwright and GPT-4o, scores candidates with OpenAI embeddings and cosine similarity, and returns ranked shortlists with LLM-generated assessments, cutting manual screening time 90%+ while processing up to 50 profiles per role for under $0.01 in API cost.

Student Research Intern, National Astronomical Observatory of Japan, Tokyo

period

Jan 2025 – Dec 2025

• Designed and trained Attention-based U-Net architectures that achieved a 38–41% reduction in reconstruction-induced parameter uncertainty relative to interpolation methods and baseline neural models, validated on a large-scale dataset of 521 gamma-ray bursts.
• Designed training pipelines, applied extensive hyperparameter tuning, and explored uncertainty quantification and ensemble methods to improve stability, interpretability, and generalization for astrophysical parameter estimation, contributing to two international research publications.

SparQ Summer Intern, QNu Labs, Bengaluru

period

May 2025 – Jul 2025

• Designed and deployed highly available infrastructure on Kubernetes with Docker and CI/CD pipelines, improving scalability, fault tolerance, and reliability of cloud-native service delivery within the organization.

A.3 Publications ¶

1. [1] Manchanda, A., Kaushal, A., Dainotti, M. G., Gupta, K., Deepu, A., Naqi, S., Felix, J., Indoriya, N., et al. “Gamma-Ray Burst Light-curve Reconstruction: A Comparative Machine and Deep Learning Analysis.” The Astrophysical Journal Supplement Series, 281:35, December 2025. Open access. doi:10.3847/1538-4365/ade851
2. [2] Kaushal, A., Manchanda, A., Dainotti, M. G., Gupta, K., Nogala, Z., Madhan, A., Naqi, S., Kumar, R., Oad, V., Indoriya, N., et al. “Multi-Model Framework for Reconstructing Gamma-Ray Burst Light Curves.” Journal of High Energy Astrophysics, 51:100519, March 2026. doi:10.1016/j.jheap.2025.100519

A.4 Selected Projects ¶

nikunjindoriya.com

period

Aug 2026

stack

Astro · LaTeX · no client framework

write-up

WN-004, A Website With Rules

• Designed and built a static personal site as a numbered series of technical documents, with a single source of CV content rendered into both a web page and a pdflatex-compiled one-page resume so that the two cannot disagree.
• Enforced the constraints in the build rather than by habit: a hash guard that fails deployment when the committed PDF falls behind its sources, and an audit that gates every build on link resolution, heading order, text alternatives, and colour contrast in both themes.

Agentic Justice

period

Jan 2026 – May 2026

stack

Python · LangGraph · Llama 3.3 / GPT-4o · Qdrant · BM25

funding

Student Innovation Grant

• Engineered a 4-agent retrieval-augmented pipeline over 7,500 Indian Supreme Court judgments (93,927 embedded chunks), fusing dense (bge-large), BM25, and citation-graph retrieval via Reciprocal Rank Fusion to lift retrieval MRR by 15 points (0.79 to 0.91).
• Designed a novel 3-level citation-verification agent (existence, metadata, and semantic-entailment checks) that eliminated LLM citation hallucinations entirely (0.0% across 237 citations on 50 benchmark questions), raising citation precision by 18 points over raw retrieval.

Faculty Outreach Automation (CLI)

period

Feb 2026 – May 2026

stack

Python · LiteLLM · Playwright · SQLite

• Engineered a multi-strategy faculty-page scraper (CSS, heuristic, and LLM extraction with Playwright for JS-rendered pages) that auto-scores and ranks candidate fields, parsing names, emails, and research interests across heterogeneous layouts.
• Built the full outreach loop end to end: DuckDuckGo discovery, resume-parsed student profile, research-overlap scoring, 3-variant cold-email generation, and Gmail OAuth2 sending with SQLite-backed reply tracking and 30-day auto-timeout.

Full write-ups are planned for the engineering notes.

A.5 Technical Skills ¶

A.6 Achievements ¶

• Secured a Student Innovation Grant to design and develop Agentic Justice, a retrieval-augmented generation system for legal research.
• Secured an All-India Rank of around 7,000 (top ~7% of 0.1M+ candidates) in GATE 2026, Computer Science & IT.
• Served on IISER Bhopal's Student Activity Council Election Committee, conducting university-wide student elections for two consecutive years.

A.7 Contact ¶

email

nikunjindoriya@gmail.com

orcid

0009-0005-5663-4728

scholar

Google Scholar

code

github.com/nikunj-indoriya

profile

linkedin.com/in/nikunj-indoriya

timezone

Asia/Kolkata (UTC+05:30)

A one-page PDF of this record, in the standard resume format, is at /cv.pdf. It is generated from the same content as this page, so the two cannot disagree.