AI Engineer

Open to full-time

Bio

Abdullah L. is an AI Engineer with several years of experience in AI automation and machine learning. He has a strong foundation in developing intelligent systems and full-stack applications. Abdullah utilizes technologies such as Python, FastAPI, Node.js, React, and MongoDB to build APIs, intelligent chatbots, and automated ETL workflows. He is proficient in using tools like Docker, AWS, and Selenium, and follows best practices in real-time data processing and scalable architecture. Notably, Abdullah developed an automated financial data extraction pipeline using OCR and NLP, enhancing data accessibility and decision-making speed. He also created a restaurant recommendation chatbot with Pinecone and Groq, achieving high-speed contextual retrieval and dynamic recommendations. Abdullah holds a Bachelor's degree in Computer Science from FAST NUCES. He is well-suited for roles that involve AI-driven backend development, real-time data processing, and advanced automation projects.

Expertise

PythonNode.jsMongoDBSQLAWSSeleniumReactGitFast APIWeb ScrapingPyTorchJava

Video Intro

Assessments

General Project Assessment

(86/100)

Great on Implement LRU Cache AlgorithmMedium
Great on Understanding Rate Limiting in APIsMedium
Great on Ensuring Real-time Data Accuracy in Mobile AppMedium

Projects

Automated Financial Data Extraction & Analysis

Developed an end-to-end automated pipeline for extracting, cleaning, and digitizing structured & scanned financial reports using OCR and NLP. Implemented sentiment analysis, data validation, and key-metric extraction to assist financial teams in rapid decision-making. Built an interactive financial query chatbot with real-time responses, improving data accessibility and analysis speed.

Restaurant RAG Chatbot

Built a high-speed restaurant recommendation system using Pinecone vector search and Groq LPU-based inference. Implemented conversational memory, contextual retrieval, and dynamic menu-based recommendations. Designed a clean, responsive Streamlit UI with FastAPI backend for seamless real-time user interaction.

Product Launch Intelligence Agent

Designed a multi-agent system capable of conducting competitive research, trend monitoring, and summarizing market insights. Automated generation of product launch strategy documents, marketing content, and risk assessments. Used agentic workflows to coordinate research, planning, content creation, and reporting.

Automated Legal Web Scraping Pipeline

Engineered a scalable ETL workflow in n8n to monitor 50+ client websites, extracting unstructured HTML data and converting it into Markdown for downstream processing. Eliminated hours of manual data entry by deploying a daily-triggered automation that captures key metadata (author, date, category) within minutes, reliably parsing diverse website structures. Architected a scalable solution enabling rapid onboarding of new clients without code changes, using intelligent deduplication to filter existing records and deliver verified updates via Slack.

Experience

AI/ML Intern
Media@MarsonOct 2025 — Nov 2025
- Engineered browser automation workflows for data extraction, web navigation, and business task automation, reducing manual workload by 100%.
- Built and deployed lead generation automations to streamline prospect identification and qualification.
- Implemented scalable automations using Python, Playwright, n8n, and third-party API integrations, enhancing process efficiency by 100%.
- Handled client communication, requirement gathering, and end-to-end solution delivery independently, ensuring seamless project execution and client satisfaction.
PythonPlaywrightn8nZapier
AI/ML Intern
Stixor TechnologiesJun 2025 — Jul 2025
- Developed a universal AI meeting assistant capable of autonomously joining Zoom, Google Meet, and Microsoft Teams, enabling real-time participation across 3 major conferencing platforms using the Recall API.
- Integrated OpenAI, ElevenLabs, and webhook-based audio streaming, handling live transcription, LLM inference, and speech synthesis, supporting continuous multi-modal conversations with sub-second audio playback latency.
- Optimized streaming and response pipelines, reducing end-to-end conversational latency by ~40–50%, achieving near real-time responses (2-3s) during live meetings.
WebSocketFastAPIElevenLabsOpenAIDockerGithub
AI/ML Intern
UrduXJun 2024 — Jul 2024
- Built an AI restaurant recommendation assistant on foodpanda Data with conversational memory and accurate contextual retrieval.
- Developed a RAG pipeline using Groq for ultra-fast generation and Pinecone for precise retrieval of restaurant menus, reviews, and offers.
- Performed data indexing and embedding optimization for improved semantic search accuracy.
- Designed and deployed a clean, interactive user experience using Streamlit and FastAPI for seamless real-time conversations.
GroqPineconeStreamlitFastAPISelenium