
AI Engineer with 2+ years in Statistical Modeling & Large Language Models
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI/ML Engineer with experience in statistical modeling, machine learning, and large language models, including fine-tuning, evaluation, RAG-based systems, and edge deployment. Builds production-grade ML systems using PyTorch, Hugging Face, LangChain, and the Anthropic Claude API across NLP, computer vision, and financial analytics domains. Comfortable shipping end-to-end ML features into web services and BI dashboards.
Lahore University of Management Sciences (LUMS)
MS Artificial Intelligence · Artificial Intelligence
September 1, 2024 – Present
University of the Punjab
BS Statistics · Statistics
October 1, 2019 – September 1, 2023
LUMS Faculty Spin-off Venture
AI/ML Engineer (Part-Time)
March 1, 2024 – October 1, 2024
India
Pakistan Bureau of Statistics
Intern
January 1, 2022 – December 31, 2022
India
Quantum Feature Maps for PSX Equities
September 1, 2024 – Present
• First quantum ML study on Pakistan Stock Exchange equity data (43 stocks, 11 sectors, 13.6 years, 107,877 stock-day events). • Built a 6-phase reproducible pipeline (PennyLane, scikit-learn) with a 16-qubit Heisenberg-ansatz PQFM, sector-stratified k-NN event matching, and walk-forward backtesting producing ~1.2M out-of-sample predictions with block-bootstrap 95% CIs.
FintelliSense B2B LLM-Augmented Financial Forecasting Platform
January 1, 2023 – February 29, 2024
• Built an enterprise-grade web application for institutional financial forecasting, combining Successive Variational Mode Decomposition (SVMD) with per-mode LSTM networks (64→32→Dense(1)) for multi-asset price prediction with 95% confidence intervals. • Integrated the Anthropic Claude API as an analyst-assistant layer that converts numerical fore-cast outputs (RMSE, MAE, R2, Sharpe, Max Drawdown, anomaly rate) into client-ready natural-language commentary and exportable PDF/CSV reports. • Implemented portfolio intelligence layer supporting up to 5 simultaneous assets with rolling Z-score anomaly detection (window=30, threshold=2.5σ), Sharpe Ratio, annualized return, volatility, and max drawdown computation. • Documented stack (TensorFlow, vmdpy, statsmodels, scikit-learn) and produced a technical scala-bility report covering computational, data, concurrency, and infrastructure dimensions.
BADRI Motor Analytics Suite - ML on Insurance Portfolio (KSA)
January 1, 2023 – December 31, 2023
• Built a multi-page analytics suite for a motor comprehensive insurance portfolio (Premium and Claims datasets) targeted at an actuarial consultancy serving the GCC market. • Engineered an Executive Dashboard, Portfolio Analysis (demographics, vehicle profile, NCD, repair type), Claims Analytics (frequency, severity, development triangles), and Pricing Insights (rate adequacy, relativities, interactive target loss ratio). • Trained a Gradient Boosting model exposed as a real-time Pure Premium Calculator with live inputs, plus a Risk Segmentation page with an adjustable risk-flag threshold and high-risk segment maps. • Demonstrates the ability to analyze large insurance datasets using ML models to interpret real-world patterns and translate them into underwriting and pricing decisions via dashboards.
Claude-Powered Document Intelligence Service
January 1, 2023 – December 31, 2023
• Built a document-intelligence microservice that ingests PDF / DOCX files, chunks and embeds them, and answers structured queries via the Anthropic Claude API with citation-style references back to source spans. • Implemented tool-use / function-calling patterns so Claude can invoke internal Python tools (SQL query, calculator, web search) within a single agentic workflow, with intermediate tool outputs persisted for auditability. • Added a streaming FastAPI endpoint plus a frontend for end-user interaction; instrumented with token-usage and latency metrics for observability.
Edge AI Deployment - LLAMA 7B Compression & Inference
January 1, 2023 – December 31, 2023
• Compressed and deployed LLAMA 7B using quantization, knowledge distillation, and LLaMA.cpp; achieved >60% memory footprint reduction with low-latency inference on Raspberry Pi. • Applied Quantization-Aware Training (QAT) and K-Means quantization; profiled accuracy vs. la-tency vs. memory trade-offs across configurations.
Agentic AI & LLM Systems Development
January 1, 2023 – December 31, 2023
• Designed Agentic AI workflows with tool usage, persistent memory, and multi-step reasoning using LangChain and the Anthropic Claude API. • Fine-tuned and evaluated LLMs for NLP tasks including sentiment analysis, NER, and domain adaptation; built quantized RAG chatbots and quantified accuracy-latency trade-offs.
Cultural Fit Analysis
The candidate's project portfolio demonstrates a diverse range of applications for AI/ML, from financial forecasting and insurance analytics to document intelligence and quantum machine learning. This breadth of interest and application, combined with experience in both academic spin-off ventures and personal projects, suggests an adaptable and curious individual. The focus on real-world problem-solving and deploying solutions aligns well with a results-oriented culture. The ongoing MS in AI further indicates a commitment to continuous learning and staying current with advancements in the field.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a strong ability to translate complex technical concepts into business value (e.g., client-ready natural-language commentary, underwriting decisions via dashboards). The experience with documenting technical stacks and producing scalability reports suggests good operational awareness and communication skills for technical audiences. The volunteer experience, though brief, hints at organizational and leadership potential.