"I don't just wrap APIs. I build reliable AI systems, optimize LLM inference, and engineer production ready agent infrastructure."
I am an AI & Research Engineer bridging the gap between SOTA LLM research and hardened production infrastructure. With 3+ years of experience, I specialize in Agentic Systems, RLVR (Rule Based Reinforcement Learning), and GPU optimized Inference.
From building multi agent workflows that reduce failures to fine tuning opensource LLMs while cutting training costs, I build end to end AI systems designed for production.
| 🧠 AI & Research | ⚙️ Systems & Backend | 🛡️ Quality & MLOps |
|---|---|---|
| LLM Agents & MCP Tooling | FastAPI, WebSockets, AsyncIO | PyTest, CI/CD |
| GRPO / RLVR (TRL), LoRA | Docker, RunPod Serverless, AWS | OpenTelemetry, SigNoz Tracing |
| RAG, Agent Memory, Red Teaming | CUDA Graphs, bfloat16, KV Cache | Ruff, Mypy, Execution based Eval Harnesses |
| PyTorch, Transformers, Whisper | PostgreSQL, Redis, S3 | GitLab CI/CD, Helm, Langfuse |
|
Reinforcement Learning Project Explore my Pacman Reinforcement Learning project. View Repository → |
Agent Systems & Security Explore my AgentArena repository, focused on agent based AI development. View Repository → |
|
Systems Level TTS Engine Up to 4.5× faster inference (RTF 0.59 → 0.13) via bucketed CUDA Graphs, bfloat16, and KV-cache tuning. Zero-hallucination pipeline deployed to RunPod. |
Verified RL Training Stack Built the complete training + evaluation stack for a math LLM (GRPO on TRL). 22,796 deduplicated samples, oracle-verified gold solutions, sandbox pool sustaining 389 exec/s. 💰 Trained for < $100 |
AI Engineer (Working Student) @ Zapdeck.AI GmbH (Berlin, Germany)
Jun 2025 – Jun 2026
- Built and deployed FastAPI microservices and LangGraph multi agent workflows on AWS EKS using PostgreSQL, Redis, Docker, and Helm.
- Optimized vLLM inference and Qdrant RAG pipelines with quantization, hybrid search, and reranking.
- Implemented Langfuse observability and secure FastMCP connectors, reducing workflow failures by 22%.
Machine Learning Engineer (Full Time) @ Lyceumerce Pvt. Ltd. (Islamabad, Pakistan)
Mar 2023 – Feb 2025
- Deployed and monitored production ML and computer vision workloads on AWS.
- Fine tuned LLaMA and T5 using QLoRA, reducing training costs by 35%.
- Built RAG based conversational AI, persistent session memory, and semantic search pipelines.
M.Sc. Artificial Intelligence
Brandenburg University of Technology (BTU Cottbus–Senftenberg), Germany
March 2025 – Present
B.Sc. Computer Science
COMSATS University Islamabad, Pakistan
2019 – 2023
- 🥇 1st Place — Google Developer Agent Hackathon
- 🥈 2nd Place — Synthetic Data AI Agents and OpenEnv Challenge, hosted by AMD, PyTorch, and Unsloth
- 🏅 Finalist — Huawei Nuremberg Hackathon (2025)
- ☁️ AWS Certified Solutions Architect – Associate
- ☁️ Microsoft Certified Azure Data Scientist Associate
- 🧠 DeepLearning.AI Deep Learning Specialization
- 🐍 PCAP Certified Associate Python Programmer
I'm open to opportunities and collaborations in AI Infrastructure, Agent Systems, LLM Serving, and Applied AI Engineering.
