Available to hire
I’m Muhammad Abdullah, a Full Stack AI Engineer specializing in agentic AI, LLM orchestration, RAG pipelines, fine-tuning, and RLHF. I design autonomous AI workflows and real-time automation platforms, building scalable backend architectures with FastAPI, Flask, LangGraph, LangChain, n8n, and cloud-native services on AWS and GCP.
I’m passionate about delivering production-grade AI solutions that cut manual effort by 30-50%, improve accuracy, and enable complex task automation across healthcare, finance, marketing, and trading. I thrive on collaborating with cross-functional teams to ship reliable systems and iterate quickly to meet domain needs.
Skills
Experience Level
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Work Experience
Machine Learning Engineer at Turing
July 1, 2025 - October 28, 2025Fine-tuned Gemini-based models for OS-level task automation, enabling mobile control such as parsing WhatsApp updates to auto-reschedule calendar events. Led supervised fine-tuning (SFT) and RLHF pipelines; created task-specific datasets and refined reward models for improved contextual understanding. Designed evaluation strategies (Evals) to benchmark responses, ensuring high alignment, accuracy, and clarity across diverse user queries. Collaborated with cross-functional teams to optimize training workflows and enhance model reliability in real-world automation use cases.
Machine Learning Engineer at Motive
April 1, 2025 - April 1, 2025Created modular workflows for streamlining internal product pipelines, improving development efficiency by 20% and reducing deployment errors. Specialized in developing and deploying advanced NLP systems using GPT, BERT, and Llama, focusing on domain-specific fine-tuning and high-performance language applications. Led end-to-end pipelines for training, evaluating, and deploying LLMs using Hugging Face Transformers, PyTorch, TensorFlow, and LangChain, enabling real-time use cases like intelligent search, summarization, and content generation. Integrated vector databases (Pinecone, FAISS) for semantic search and hybrid retrieval pipelines, reducing retrieval latency by 30% while improving relevance scores in RAG applications. Authored API documentation and technical runbooks for internal and client use, cutting onboarding time for new developers by 25%. Maintained continuous learning loops and MLOps workflows using MLflow, Docker, and Git-based CI/CD pipelines to ensure reproducibility,
Software Engineer at Arbisoft Pakistan
February 1, 2023 - February 1, 2023Engineered microservices architecture on AWS with Flask backend, implementing JWT-based authorization and improving authentication reliability by 30%. Integrated Stripe for payment processing, enabling seamless subscription management and achieving 25% faster transaction flow. Designed Firebase-based synchronization system for multi-location content delivery, reducing sync latency by 40% across 50+ devices. Developed and deployed an AI-powered chatbot integrated with Pinecone, Gemini, and OpenAI embeddings, improving contextual query resolution accuracy by 35%. Led backend team of 5 developers, managing architecture decisions, code reviews, and delivery timelines, increasing on-time delivery rate to 95%. Created backend documentation and API specifications, enabling cross-functional collaboration and reducing onboarding time for new developers by 20%.
Associate Software Engineer at Devsinc Pakistan
December 1, 2022 - December 1, 2022Built and deployed backend applications using Flask and Django, strengthening expertise in REST API development and scalable server-side architectures. Automated recurring workflows and data processing tasks with Python scripting, reducing manual overhead by 30%. Worked with structured (PostgreSQL, MySQL) and unstructured databases (MongoDB, Firebase), gaining hands-on experience in data modeling, indexing, and query optimization. Implemented authentication and authorization systems, including JWT-based access control, laying the foundation for secure microservices development. Collaborated in Agile teams, actively participating in code reviews, sprints, and client demos to sharpen communication and problem-solving skills.
AI Engineer at Turing (Client: Google)
February 1, 2025 - PresentEmbedded as an AI Engineer for Google's engagement via Turing. Validated, optimized, and debugged large-scale production AI codebases to improve model correctness, performance, and reliability. Designed prompt engineering strategies for LLM-based workflows, created structured prompts, constraints, and evaluation tests to improve reasoning and output consistency. Implemented prompt iteration and RLHF workflows to align model behavior with human feedback and domain requirements.
AI Engineer at Nexxt.ai
July 1, 2023 - January 1, 2025Architected and maintained secure, scalable AI-powered platforms across Fintech, E‑commerce and Real Estate focusing on compliance, real-time processing, and service resilience. Built an AI-driven voice agent handling 20+ concurrent calls with sub-1000 ms end-to-end latency; integrated real-time speech processing and backend API checks for availability and scheduling. Implemented ML systems including F1 outcome predictor, crypto market trend classifier, NLP resume parsing, and image-based product search; productionized with FastAPI and AWS SageMaker with versioning, retraining triggers, logging, and monitoring.
Machine Learning Engineer at Ascendia AI
January 1, 2022 - June 1, 2023Developed AI-driven recruitment platforms with real-time voice AI, RAG-based semantic search, and event-driven microservices. Implemented LiveKit-based Voice AI interview system, parallel crawlers across LinkedIn/Indeed/Dice, RAG-based semantic matching with LED-based long-document summarization and OpenAI embeddings with Pinecone, adaptive conversational flows with LangGraph, and CPU-optimized fine-tuned models via 8-bit quantization and ONNX Runtime. Built an AI phone outreach system with Twilio, reducing candidate response time.
Software Engineer at Infostack Pakistan
February 1, 2019 - December 1, 2021Backend engineer for healthcare and financial platforms, building secure APIs, event-driven services, and data pipelines; implemented JWT/OAuth 2.0 with RBAC, rate limiting, and Stripe integration with idempotency and retries. Optimized performance with Redis caching, query optimization, and database indexing. Designed PostgreSQL and MongoDB schemas; deployed on AWS and GCP with Lambda/EC2/Cloud Run, auto-scaling, logging, and monitoring. Automated compliance data collection via web scraping projects using Selenium and Scrapy.
Education
Bachelor of Science in Computer Science at University of the Punjab
January 1, 2017 - January 1, 2021Bachelor of Science in Computer Science at University of the Punjab
January 1, 2015 - January 1, 2019Qualifications
IBM RAG and Agentic AI Specialization
January 1, 2025 - April 16, 2026Industry Experience
Software & Internet, Professional Services, Computers & Electronics, Media & Entertainment, Healthcare, Financial Services
Hire a Full Stack Developer
We have the best full stack developer experts on Twine. Hire a full stack developer in Lahore today.