Yo! I'm Ankit
Product Manager by profession, Entrepreneur at heart. ❤️
I like building things people actually use — not just ship features and move on, but figure out why something is broken and fix the actual problem. That's basically what I have worked on so far.
Right now I'm Working at Fitelo as an Associate Product Manager in Founder's Office, building EZWeight — a doctor-led GLP-1 weight loss platform.
















My Startup Experience
Started a food-tech D2C brand at 18. Scaled to Decent revenue and Served 1400+ customers. 🚀
How I Built a Food-Tech Startup From Scratch: Problem → Product
As founder and product lead, I owned discovery, roadmap, and metrics — validated demand before writing production code, then built the ordering platform in Next.js and ran sourcing, pricing, and logistics end-to-end.
Where I've Worked
3 internships across Fitelo, AICTE, and Fretbox — owning product experiences and shipping AI features to hundreds of thousands of users.
Founder's Office at Fitelo
Associate Product Manager
Aug 2026 – Present · Gurugram, HaryanaWorking on EZWeight, Fitelo's health and wellness app — from discovery through design, tech, strategy, and business teams to prioritize the roadmap and translate business goals into deliverables.
Collaborated with designers on UX flows and engineering on technical feasibility, driving alignment across teams to ship features on scope, on time, and to a high quality bar.
Management Trainee
Feb 2026 – Aug 2026 · Internship- Built a RAG-based Persona Engine delivering personalized health recommendations across Fitelo's 2 lakh+ users, combining vector search and LLM inference to lift engagement.
- Architected AI pipelines integrating RAG with LLM orchestration to auto-generate personalized diet plans at scale, cutting manual touchpoints.
- Built multi-agent n8n workflows to autonomously manage health journeys — triggering interventions and syncing data across services.
Product Management Intern at AICTE
Hyped Samvadini — Delhi HQ
Dec 2024 – Jan 2025 · New Delhi, India- Contributed to product strategy for Hyped Samvadini, a government-backed messenger built around secure chats, calls, and video — with end-to-end encryption and a privacy-first identity model (hidden number + Dynamic Chat ID).
- Defined requirements for real-time video calling and anti-spam request-based messaging, ensuring a safe, reliable cross-platform (Android/iOS) experience.
- Shaped core user flows — onboarding, contacts, messaging — supporting the app’s multilingual (55+ languages) and inclusive design for Bharat’s users.
AI Product Intern at FretBox
Binary Technologies
Aug 2024 – Nov 2024 · Remote, Noida- Contributed to product development of FretBox’s Resident App, covering key user journeys — onboarding, room booking, rent payments, and complaint management — for a platform now used across 55+ campuses and 80,000+ students in India.
- Defined requirements for real-time features like room availability, payment status, and maintenance request tracking, working closely with engineering to replace fragmented WhatsApp-based communication with a unified in-app experience.
- Led product exploration for an LLM-powered chatbot (OpenAI API) to automate resident query handling and hostel notifications — reducing repetitive support load on wardens and admins.
Projects
Things I've built and shipped.
Unhinge
Locality-First Social App for Real-Life Friendships
- Designed and built Unhinge, a locality-first social app helping 18–25 year-olds in Gurugram form real-life friend groups — human-reviewed profiles, no swiping, no DMs.
- Defined the core product loop: users express interest in their locality → once 7+ people opt in, a WhatsApp group auto-forms — replacing algorithmic matching with a queue-and-threshold model.
- Built trust and safety into the product from day one — fuzzed location display (locality-only, no exact address), phone number privacy, and a reporting system for every profile.
Zynri
Delhi NCR Startup Map & Directory
- Built Zynri, an interactive map and directory of the Delhi NCR startup ecosystem, letting users discover startups by location, stage, sector, and type via map or grid view.
- Designed the discovery experience — filterable search (area/stage/sector), a “Near Me” view, and a public submission flow for founders to list their startups.
- Introduced a monetization layer with a “Boost Pin” feature, letting startups pay to increase visibility on the map.
About Me
I’m a curious and easy-going guy who loves meeting new people, exploring new places, having random deep conversations, and trying things I’ve never done before. I’m pretty spontaneous, always up for an adventure, and probably overthink things more than I should. 😭
Beyond that, I genuinely care about society and how we can make it a little better. I enjoy taking part in social initiatives, volunteering with different organizations, and contributing to causes that create a real impact. I like being involved, meeting people from different backgrounds, and doing my bit rather than just talking about change.
Technical Skills
CS engineering foundation with deep product, AI, and full-stack expertise.
Product Skills
Technical Skills
AI & GenAI
Tools & Platforms
Education & Volunteering
Bachelor of Engineering
Computer Engineering. 2022 - 2026
THAPAR INSTITUTE OF ENGINEERING & TECHNOLOGY
Joint Secretary
Linux User Group (LUG), TIET
- Led an AI & GenAI Workshop Series guiding students toward open-source contributions and programs like GSoC.
- Conducted hands-on workshops on LLM Fine-tuning, Prompt Engineering, RAG Systems, Agentic AI, Git/GitHub, and Full Stack Development.
- Mentored peers on AI/ML projects, LLM applications, and autonomous agent development, fostering technical depth in emerging AI domains.
Research Projects
Exploring decentralized ML and non-invasive health diagnostics.
Federated Learning Research | AP2PFL Method
- Mentored by: Dr. Anjula Mehto and Mukta Sharma (MSU, USA); manuscript in preparation.
- Co-developed AP2PFL (Adaptive Peer-to-Peer Federated Learning), combining CNN-based client training with FedAvg aggregation for decentralized model training.
- Achieved 57% peak accuracy on non-IID distributed data, a 12% improvement over standard FedAvg baseline.
- Researched privacy-preserving ML and decentralized learning architectures suited for sensitive healthcare data.
AI-Based IOP Monitoring Device
- Developed an AI-driven intraocular pressure (IOP) monitoring system using deep learning for non-invasive, real-time eye health assessment.
- Designed a sensor-based signal processing pipeline with neural networks for glaucoma risk detection and early diagnosis.
- Filed patent (Application No. 202511115771) under TIET for acoustic excitation-based AI monitoring technology.
- Demonstrated 95% detection accuracy in preliminary trials, advancing non-invasive ophthalmic diagnostics.
A few things I'm proud of
Co-founded and scaled a D2C food-tech startup to a Decent revenue and 1400+ orders while building the full-stack ordering platform from scratch.
Built RAG pipelines for 2L+ users at Fitelo and filed a patent (202511115771) for an AI-based non-invasive glaucoma detection device.
Shipped core features for a government messaging app and an 80K-student SaaS platform, replacing fragmented flows with unified experiences.
Mentored 250+ students in AI & GenAI as Joint Secretary of TIET's Linux User Group, driving open-source contributions.
Co-founded and scaled a D2C food-tech startup to ₹8L+ revenue and 600+ orders while building the full-stack ordering platform from scratch.
Built RAG pipelines for 2L+ users at Fitelo and filed a patent (202511115771) for an AI-based non-invasive glaucoma detection device.
Shipped core features for a government messaging app and an 80K-student SaaS platform, replacing fragmented flows with unified experiences.
Mentored 500+ students in AI & GenAI as Joint Secretary of TIET's Linux User Group, driving open-source contributions.
Co-founded and scaled a D2C food-tech startup to ₹8L+ revenue and 600+ orders while building the full-stack ordering platform from scratch.
Built RAG pipelines for 2L+ users at Fitelo and filed a patent (202511115771) for an AI-based non-invasive glaucoma detection device.
Shipped core features for a government messaging app and an 80K-student SaaS platform, replacing fragmented flows with unified experiences.
Mentored 500+ students in AI & GenAI as Joint Secretary of TIET's Linux User Group, driving open-source contributions.