syed anas

Full Stack & AI Engineer specializing in React.js and Agentic Workflows

Based in: Karachi, PakistanCurrently: Full Stack & AI Engineer, StratesfyExperience: 2 years

A final-year Computer Science student and Full Stack Engineer with extensive experience building responsive web applications and AI-powered automation systems. He has successfully delivered over 10 production client UIs and architected autonomous web-scraping agents using LLM APIs and n8n.

Experience

Full Stack & AI Engineer

StratesfyRemoteworkSep 2024 – Present

Summary

Engineered 10+ production client UIs using React.js, TypeScript, and Next.js, reducing average page load time by 30% through component-level code splitting and modern CSS layout techniques. Designed and built REST API-driven backend workflows using Node.js and FastAPI, cutting manual data entry overhead by 15 hours per week across an internal CRM serving multiple client accounts. Architected an autonomous AI web-scraping agent using LLM APIs, Python, and n8n automation pipelines, increasing structured data extraction accuracy by 25% while processing 500+ GB of unstructured content. Deployed and maintained full-stack applications on Vercel and cloud infrastructure, implementing CI/CD pipelines that reduced release cycle time by 40% and eliminated manual deployment steps. Built an AI-powered SaaS landing page builder with a FastAPI backend, integrating LLM APIs to generate validated JSON content and a drag-and-drop editor using Zustand and dnd-kit, reducing API call latency by 20%. Managed sprint planning, stakeholder communication, and cross-functional coordination across 5+ concurrent client projects using Jira and Agile methodology, achieving on-time delivery for all major milestones.

What I did

Implemented structured decoding using Pydantic and OpenAI function calling with strict JSON schema.

Results

Created evaluation cases for tool calls and maintained trace logs to re-run after prompt or model changes.
Implemented automatic entry logic for instances where Pydantic validation failed.
Reduced average page load time by 30% through component-level code splitting.
Cut manual data entry overhead by 15 hours per week.
Increased structured data extraction accuracy by 25% while processing 500+ GB of content.
Reduced API call latency by 20% for SaaS landing page builder.
Achieved on-time delivery for all major milestones across 5+ concurrent client projects.

React.jsTypeScriptNext.jsNode.jsFastAPILLM APIsPythonn8nVercelCI/CDZustanddnd-kitJiraAgilePydanticOpenAI function callingJSON schemallama.cppGGUFQuantization

Machine Learning Intern

Nodebook Pvt LtdKarachiinternship2023 – 2023

Summary

Accelerated model iteration speed by 35% by engineering reusable data preprocessing pipelines in Python using Pandas and Scikit-learn, covering EDA, feature engineering, and automated validation checks. Improved model test coverage by authoring structured unit test suites for ML pipelines, catching 90% of data-quality issues before production evaluation runs. Contributed to 3 end-to-end ML model builds from data ingestion to evaluation, collaborating with senior engineers to document findings and propose performance improvements.

Results

Performed manual and regression testing using Pydantic validation and a golden dataset for data pipelines to catch schema drift.
Caught 90% of data-quality issues before production evaluation runs.

PythonPandasScikit-learnEDAPydantic

Skills

Technical

TypeScript

Next.js

JavaScript

HTML

CSS

Tailwind

React.js

Prompt Engineering

Python

LLM APIs

Git

GitHub

Vercel

n8n

Manual Testing

Regression Testing

Pandas

Test Case Design

MongoDB

MySQL

C++

Assembly

PostgreSQL

FastAPI

Zustand

Pydantic

Data Structures

Object-Oriented Programming

Artificial Intelligence

Human Computer Interaction

Structured Decoding

Outlines

Ollama

llama.cpp

llama-cpp-python

Model Quantization

Ollama Modelfile

llama.cpp quantize

AWQ

Projects

AI Web Scraping Agent

AI Engineerpersonal

Built an AI agent to extract and structure website data. Automated workflows using n8n. Improved accuracy with prompt-based logic.

n8nLLM APIsPython

Investigation Game Agentic Coding

Developerpersonal

Built a decision-based interactive game using AI logic. Implemented branching decision systems.

AI LogicPython

Restaurant Website (React.js)

Frontend Developerpersonal

Developed a responsive restaurant website with dynamic UI. Used React Router and reusable components.

React.jsReact Router

Synova Pharma Website

Frontend Developerprofessional

Built a professional pharma website with responsive design. Structured business-focused UI sections. Delivered with high client satisfaction.

Responsive DesignUI/UX

E-Book Management System

Developerpersonal

Developed a digital library system with search and filtering. Used JavaScript for dynamic DOM updates. Designed a clean and user-friendly interface.

JavaScriptDOM Manipulation

Brikline

personal

A SaaS platform for homebuilders. Engineered the backend to use PostgreSQL for primary data storage and MySQL specifically for user account management.

PydanticOpenAI function callingJSON schemaPostgreSQLMySQL

SaaS Landing Page Builder (AI-Powered)

Full Stack Developerpersonal

Built a full-stack AI SaaS that generates landing pages from product descriptions. Developed drag-and-drop editor with Zustand state management. Implemented authentication, CRUD, auto-save, and publishing. Integrated LLM APIs with FastAPI backend.

FastAPILLM APIsZustandReact.jsPydanticOpenAI function callingJSON schemaLlama 3 8BMistral 7BGGUFllama.cppOllamaModelfilevLLMAWQ

Education

BS Computer Science

Iqra UniversityComputer Science2022 – 2026

Intermediate

DJ Govt Science CollegePre-Engineering2020 – 2022

Hire syedReach out about a role, a contract or a conversation.For recruiters

syed's twin is AI, it can make mistakes.

More people on Kanvis

Browse more people Create your own Kanvis