Markham, Ontario, Canada
Aatif Usmani
AI Engineer at Tasteport, one of the first 10 members of a VC-backed startup. I build AI agents, assistants, and RAG systems in production.
▸ press start · choose your class ◂
AI Engineer at Tasteport, one of the first 10 members of a VC-backed startup. I build AI agents, assistants, and RAG systems in production.
▸ press start · choose your class ◂
I build AI products at Tasteport, a VC-backed ethnic grocery platform where I was one of the first 10 employees. My day-to-day is architecting Mina, our AI customer shopping assistant.
I also do hackathons and other projects that I find interesting on the side. You can check out what I've built in the projects section below.
> init aatif_usmani --role
AI Engineer
Tasteport · Toronto, ON
6+
Years at Tasteport
location ... Markham, ON
email ...... ventures.aatif@gmail.com
github .... /AatifUsmani
education . B.Sc. Biochemistry, CS minor · transitioned to full-time at Tasteport
The system uses a LangGraph agent with tool calling to search product data and return relevant product cards, while RAG enables the assistant to retrieve information from FAQ and support documentation. I implemented semantic long-term memory using profile-based vector storage, allowing Mina to retain relevant information from previous interactions.
Reduces redundant LLM API calls, using vector similarity comparison to reduce token consumption and lower API costs. Implemented a middleware layer that converts inbound text queries into semantic vectors using the OpenAI API. Utilized a NumPy-driven matrix comparison pipeline with custom cosine similarity thresholds to establish baseline cache accuracy. Integrated basic Time-To-Live (TTL) mechanics and data validation checks to prevent serving stale or outdated cached responses.
1st Place – YorkU Data Storytelling Competition. Built an anomaly detection dashboard utilizing an Isolation Forest ensemble and domain-specific rules to map toxicity-weighted exposure data over geographic regions.
Developed an automated data pipeline monitoring tool to detect schema drift and evaluate downstream column-level lineage impacts. Utilized LangGraph with human-in-the-loop approval gates to process schema repairs, validating outputs against strict Pydantic specifications and static checks before generating dbt artifacts.
Operations and CX Intern
B2B, B2C & fulfillment relationships
Data Associate (Part-Time)
Inventory data, ops tooling & audits
AI Engineer
Production AI assistant
AI Customer Shopping Assistant
Designed, Built and Deployed Production AI Shopping Assistant - Mina | Python, FastAPI, LangGraph, LangChain, OpenAI, PostgreSQL, Pinecone, Docker, and AWS. Resolved ~70% of customer-support inquiries without human-agent escalation
Tool calling for real-time product search
Built a LangGraph tool-calling pipeline that enables the agent to query the product catalog and return structured product cards directly to customers for cart actions
Memory
Implemented short term conversational memory using LangGraph PostgreSQL checkpointers to persist agent state, alongside semantic long-term memory using LangGraph and Pinecone, isolating user-specific embeddings into separate namespaces to support personalized context
RAG
Built a RAG pipeline for FAQ and support documentation using LangChain, OpenAI embeddings, and Pinecone to retrieve semantically relevant context for customer queries
Error Handling, Schema Validation
Designed type contracts between LLM outputs and app logic using Pydantic schemas, structured outputs, and validation, preventing malformed model responses from propagating through the application. Implemented bounded retries, timeouts, validation handling, user-facing error states, and model fallbacks to maintain reliable behavior across LLM and backend failures
Observability
Implemented production observability across AI and backend infrastructure using structured application logging, AWS CloudWatch monitoring, and LangSmith traces
AI / ML
Lv.87
EXP 119190 / 99999
RAG & Agents
Lv.84
EXP 115080 / 99999
Backend & Data
Lv.78
EXP 106860 / 99999
Cloud & DevOps
Lv.74
EXP 101380 / 99999
York University
B.Sc. Biochemistry, Minor in CS · transitioned to full-time work (2025)
AI & Data Society
IT Engineer (2025)
Wilfrid Laurier
Business & Science (2022–2024)
YorkU Data Competition
1st place - Invisible City
NVIDIA
Certified Associate - Generative AI & LLMs
DeepLearning.AI
Deep Learning Specialization