# Arushi Gupta
> Agentic AI engineer — builds the systems, speaks the finance.
Role AI Fellow · LRL Finance Analytics, Eli Lilly
Base Boston, MA
Edu BS Computer Engineering, Purdue '26 (AI/ML)
Open `work.ts` to see what I've shipped — or press ⌘K.
/**
* I build and ship agentic AI — multi-agent systems that do
* real work, with computer-engineering fundamentals underneath
* and a finance domain on top. The overlap is where I'm useful:
* the deploy pipeline, the math, the layer below the API.
*/
export const arushi = {
role: "AI Fellow",
team: "LRL Finance Analytics, Eli Lilly",
base: "Boston, MA",
edu: "BS Computer Engineering, Purdue '26",
focus: ["agentic AI", "systems", "finance"],
likes: "the parts most people skip",
};
type Project = { name: string; stack: string[]; note: string };
export const work: Project[] = [
{
name: "graphrag-rust",
stack: ["Rust", "GraphRAG", "Retrieval"],
note: "KG retrieval vs. vanilla vector RAG — in Rust, below the API.",
},
{
name: "news-intelligence-pipeline",
stack: ["GCP", "Pub/Sub", "LLM"],
note: "Live news in, structured intelligence out — at stream scale.",
},
{
name: "outbound-sales-agent",
stack: ["Multi-agent", "LLM", "Telephony"],
note: "Qualifies a lead, then routes: close / book / nurture. @ ReVenture.",
},
{
name: "malicious-activity-predictor",
stack: ["Data mining", "Anomaly detection"],
note: "Flags threats in high-dimensional data before they escalate.",
},
];
// also: NNs & decision trees from scratch · Duckietown self-driving
// (ROS + PID + CV) · speech-to-text on a Pi · OCR receipt → budget
[
{
"org": "Eli Lilly",
"role": "AI Fellow — LRL Finance Analytics",
"years": "2026—",
"what": "Build & deploy multi-agent systems for financial analysis; own the path to prod."
},
{
"org": "ReVenture AI",
"role": "AI Engineer (led intern team)",
"years": "2025",
"what": "Led the outbound sales agent — qualification + four-way routing, CRM + telephony."
},
{
"org": "Outlier AI",
"role": "Generative AI Trainer",
"years": "2024",
"what": "Ranked LLM outputs on math & Python (RLHF-style) to improve quality."
},
{
"org": "Purdue University",
"role": "TA — ECE 368, Data Structures & Algorithms",
"years": "2025",
"what": "Mentored students through trees, graphs, DP & memory management in C."
}
]
export const stack = {
languages: ["Python", "Rust", "C/C++", "SQL", "TypeScript"],
agenticAI: ["multi-agent", "RAG/GraphRAG", "prompt eng", "RLHF eval"],
ml: ["neural nets", "decision trees", "CV", "data mining", "NLP"],
dataCloud: ["GCP (Pub/Sub, Scheduler)", "Microsoft Fabric", "Denodo"],
ship: ["GitHub Actions / CI-CD", "Docker", "OIDC"],
financeBI: ["Power BI", "FactSet", "PitchBook", "variance decomp"],
};
# Get in touch
Let's build something that ships.
- email arushigupt.23@gmail.com
- github github.com/i4rushi
- linkedin linkedin.com/in/arushig05
<!-- résumé in the sidebar: resume.pdf -->