Background
About Me
Software engineer with hands-on experience building AI-driven backend systems and automation pipelines used in production. I've worked on LLM-powered agents, RAG pipelines, and API-integrated workflows, with growing exposure to MLOps practices such as experiment tracking, data versioning, and CI/CD.
Currently focused on bridging experimental AI systems with reliable, maintainable production code. My work spans from designing LLM-based multi-agent systems to building end-to-end ML pipelines with proper versioning, tracking, and deployment automation. I'm passionate about making AI systems production-ready and scalable.
Education
BE Software Engineering
National University of Sciences & Technology (NUST)
Expected June 2026
CGPA: 3.43 / 4.00
Relevant Coursework
Experience
Professional Work
AI Intern
ANTEMATTER
- Contributed to a multi-agent email automation system that now automates 90% of bookings for a global black-car client operating in 40+ countries, serving 60% of the Fortune 100.
- Designed deterministic prompt templates, intent classifiers, and extraction schemas while integrating LLM components with private booking APIs via n8n multi-workflow HTTP calls to automate client requests and escalate edge cases to human agents.
- Helped engineer, validate, and harden a reusable AI agent framework through testing, documentation, and schema enforcement, improving output consistency and reliability in production workflows.
Projects
Selected Work
Churn Intelligence Platform
MCP-Powered MLOps System with LLM Tooling
F1 RAG Assistant
Hybrid Retrieval-Augmented Generation for Formula 1
Federated Healthcare Data Platform
Decentralized Analytics with Privacy-Preserving AI Pipelines
Cloud Video Streaming Platform
Microservices Architecture on Google Cloud Run
Crypto Intelligence Dashboard
Real-Time Market Data with NLP Summarization
TechConfig — E-commerce Platform
Full-Stack Store with Custom PC Builder
← Swipe to explore →
Credentials
Research, Certifications & Achievements
Research
Horizon-Aware Label Efficiency in Energy Infrastructure Forecasting via Masked Pretraining
Submitted to Springer Nature — Under Review
Developed a compact masked-pretraining framework (~150K parameters) for long-horizon time-series forecasting that achieved up to 95% of full-supervision accuracy using 25–50% labeled data and reduced MAE by 8–14% on ETTh benchmarks through variance-weighted masking and leakage-free evaluation.
View PaperCertifications
Machine Learning Specialization
DeepLearning.AI & Stanford
Supervised/unsupervised learning, regression, classification, PCA, and bias–variance trade-offs.
Deep Learning Specialization
DeepLearning.AI
(In Progress: 3/5) Neural networks, backpropagation, CNNs, sequence models, and optimization.
Agentic AI
DeepLearning.AI
Agentic workflows, reflection patterns, tool use, and autonomous agent design.
Python for AI
DeepLearning.AI
AI solutions using Python — automation, data handling, and AI workflow implementation.
Programming Essentials in C++
Cisco Networking Academy
Core programming concepts, object-oriented design, and algorithmic problem-solving.
Kaggle
Intro to Machine Learning
Kaggle
Decision trees, random forests, model validation.
Intermediate Machine Learning
Kaggle
Missing values, categorical vars, pipelines, cross-validation.
Feature Engineering
Kaggle
Powerful feature creation for improved model performance.
Data Cleaning
Kaggle
Handling missing values, scaling, and data quality.
Achievements
STEP ECAT Award (3rd Place)
Ranked 3rd in Pakistan's largest engineering admission test.
Google Dev Club TAG'25 CP (4th Place)
Competitive programming competition at NUST.
Talent Award (Multiple)
Top academic performer in city-wide competition.
LeetCode 300+ Problems
Ranking in the top 3% of programmers globally.
Connect
Skills & Contact
AI / ML
Backend
MLOps & Automation
DevOps & Cloud
Languages
Let's Work Together
I'm always open to discussing new opportunities, research collaborations, or interesting projects.
© 2026 Ahmed Bilal