build · ship · iterate

Ayaan Khan

ai engineer · builder

Open to work

Chicago, IL ayaanahmedkhan12@gmail.com

GitHub ↗ LinkedIn ↗

local 02:26 AM CST

./me.jpeg

activity

building: VeritasLayer (building),TrueTone (building),This site
learning: Agentic workflows in construction and real estate.
watching: The Bear (Show, 2026) · previously Beef, Succession
reading: Attention Is All You Need (Vaswani et al.)
hardware: MacBook Pro M4 Pro · iPhone 17 Pro · Samsung Z Fold 7
software: VS Code · Claude · Codex · Gemini · Ghostty
supporting: FC Barcelona & Argentina

Recently played

// 02 — about

About Me

AI engineer building production LLM systems — RAG pipelines, semantic search, multi-agent workflows. Originally from New Delhi, based in Chicago. Outside the terminal, I'm an avid football fan (the real kind), play guitar, and spend too much time thinking about music and art. I think good taste is a technical skill.

// where I'm headed

I'm launching TrueTone this fall (truetone.skin) — a shade-matched makeup and skincare marketplace I founded, from concept to a working iOS & Android app and a live pre-launch waitlist. After that I want to keep shipping AI systems that actually matter — whether that's at a fast-moving startup or a company building infrastructure for the next wave of intelligent software. Longer term, I want to keep building products of my own at the intersection of AI and systems design, and eventually contribute to research that makes LLM pipelines more reliable and interpretable.

education

Degree: B.S. in Artificial Intelligence
Minor: Minor in Architecture
University: Illinois Institute of Technology
Completion: May 2026

// coursework

01 DSA02 AI03 ML04 NLP05 DBMS06 Assembly07 Data Mining08 Discrete Math09 Linear Algebra10 Probability11 Statistics12 OOP

// 03 — experience

Experience

TrueTone · Chicago, Illinois · 2025 – Present

Founder

1. Founded TrueTone, a shade-matched makeup & skincare marketplace launching fall 2026 — concept to a working iOS & Android app and a live pre-launch waitlist, with shade-matching as the shop-first hook.
2. Built on-device CV (React Native Vision Camera + ML Kit) that reads tone and undertone from a selfie — image never leaves the device — plus a shade-matching engine and Expo / React Native app on Supabase, designed BIPA/GIPA/FTC compliant.

on-device shade match, launching fall 2026

D.A. Syntec Ltd. · Chicago, Illinois · May 2025 – May 2026

AI & Digital Developer

1. Built and deployed the production RAG layer over firm documents (OpenAI + ChromaDB) — now the shared knowledge backbone for the SyntecAI suite, with cited sources instead of file-share hunting.
2. Built the CCN Agent: a GPT-4o function-calling agent that looks up and edits classification codes in natural language, backed by SQLite + ChromaDB with atomic writes and confirmation gates on destructive edits.

atomic cross-store writes

Excelerate (Globalshala) · New Delhi, India · June 2023 – July 2023

Project Manager

1. Led a distributed global team to organize an academic event with a $30,000 budget, owning planning through delivery across time zones.
2. Ran documentation, risk assessment, and external vendor outsourcing so the event stayed on scope and on budget.

Volunteers.Covihelp · Remote (India) · May 2021 – July 2021

Co-Founder

1. Co-founded a 24/7 helpline during India's second COVID wave, connecting thousands of patients to oxygen, beds, and medicines through personally verified leads.
2. Operated volunteer intake across WhatsApp and Messenger, ran relief fundraising, and earned coverage in Hindustan Times, Firstpost, YourStory, and other outlets.

// 04 — projects

Projects

↳ select a row for detail

// earlier work

// 05 — systems

Systems Thinking

01
Ingestion & preprocessing
Clean, chunk, and index messy docs so retrieval starts from ground truth
02
Hybrid retrieval + reranking
Vector search alone hallucinates on structured data — combine with lexical matching and rerank
03
Agentic workflows & guardrails
Function-calling agents with confirmation gates before anything destructive touches production
04
Structured outputs & validation
Every LLM response parsed through Pydantic schemas — if it doesn't conform, it doesn't ship
05
Latency/cost routing
LiteLLM across models, Redis caching, and fallback LLMs — not every query needs GPT-4o
06
Observability & failure modes
Inngest job tracing, eval harnesses — catch breaks before users do