Who is Harshit Agarwal?

Harshit Agarwal is a Fullstack Software Architect and AI Researcher based in India. He acts as Founding Engineer & Lead SDE at platforms like Famcare, shoppin', and endorphind, specializing in high-throughput backend services and deep learning systems.

What technologies does Harshit Agarwal use?

Harshit specializes in Python, FastAPI, Go, C/C++, Java, React/Next.js, Flutter, PyTorch, AWS (SageMaker, EKS, EC2), Kubernetes, Rancher, Docker, PostgreSQL, Redis, and WebSockets.

What research has Harshit published?

Harshit has published research on transformer-based NLP architectures for software vulnerability exploit prediction in NVD (AIP / Springer), and has submitted gait analysis/Parkinson's Freezing of Gait prediction research to Health and Technology.

What open-source projects has Harshit built?

Harshit is the author of gaitSetPy (a Python package for high-throughput gait time-series preprocessing and modeling) and has built custom generative video nodes for ComfyUI workflows.

shoppin' Experience

Every year of my life arranges itself into this weirdly dissonant composition. Some cadence, constantly interrupted.

Second half of 2024, I stumbled onto this Instagram ad. A girl describing a type of person everyone has in their friend group: the fashion searcher. The one who does the dirty digging. You spot a guy in Sarojini in the dopest baggy jeans, snap a pic, send it to her. She finds it.

Crowded Sarojini Nagar market with a Police Station sign and street vendor stalls — Sarojini Nagar. If you know, you know.

That was shoppin's pitch. One snap, no fuss, 10 websites with prices compared.

Snap a pic

→

shoppin' search

→

10 sites compared

Fig. 1 — the pitch: one photo in, prices from 10 websites out.

What got me wasn't the tech. It was the marketing. Unrestrained, experimental, willing to try things other fashion brands wouldn't touch. A curation with MyMuse India was the moment I thought, these people think like me. Where others see boundaries, they see opportunity.

I DMed them on Instagram, got their email, and then forgot to apply for two months.

Mid November. End sems running. Something told me, now.

I wrote the strangest application email. Barely anything about tech, almost entirely about fashion. Resume attached at the bottom like an afterthought.

Screenshot of the actual application email sent to shoppin', mentioning fashion, cosplay, and a 'find me a cosplay feature' joke — The actual email. Barely any tech, mostly fashion.

Got the assignment faster than I expected, running model training on images to build a fashion search lens. I wanted to go further, build a cosplay searcher on top: feed it a character, get matched clothes from across the internet. Asked for an extension. Submitted a report covering 10-12 image search methods with full performance metrics.

10-12

image search methods evaluated

45min

CTO interview call

11-12mo

Dec 2024 – Nov 2025

Got an interview call with the CTO who spent 45 minutes trying to convince me to take a drop instead.

Fun times.

The next 11-12 months, December 2024 to November 2025, I learnt and contributed more than I can summarize here. That deserves its own post.

Late night office desk setup with monitors showing analytics dashboards and code, a soda can, and notebooks

Dark office at night with the team gathered around a projector screen during a campaign launch, captioned 'campaign about to start, we with our last min deployments'

But when I look back, the best part was the culture. Night rides to Treat, Chanakyapuri.

Anna's The Treat cafe at night in Chanakyapuri, neon signage, people gathered outside — Treat, Chanakyapuri. Most of the real conversations happened here, not at a desk.

Work that kept me on my edge. Challenges I genuinely adored. And people beside me debating with full sincerity whether Alia Bhatt's saree-gown is a saree or a gown.

Indoor cricket net session at night, players batting and fielding under floodlights — Indoor cricket after hours. The gedi didn't stop at ludo.

Special mention to Adi G., Nishant, Lokesh, Sehaj for the gedi, late night ludo, and all the bak bak.
Suraj Bahuguna, we have more left to do in Bangalore.
Harshit Saidaniya, for the random drop that reminded me I'm not the only one among these uncs.
Manas Goel, for being an amazing coworker and mentor.
Shlok and Utsav, for the opportunity.

More on the work itself: shoppin' Case Study