AI Engineer with less than a year in LLM Applications & Computer Vision
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 Engineer specializing in LLM applications, computer vision, and ML infrastructure. Experienced in designing, deploying, and scaling end-to-end AI systems, with a focus on building reliable, production-ready solutions that deliver measurable real-world impact.
National University of Sciences & Technology (NUST)
BS · Computer Science
August 1, 2022 – June 30, 2026
Neuronix Technologies
AI Engineer Intern
March 1, 2026 – April 1, 2026
India
Data BI LLC
Data & AI Associate Engineer
June 1, 2025 – December 1, 2025
India
Voice 2 Action
June 24, 2026 – Present
End-to-end agentic system that converts informal Roman-Urdu voice notes into structured tasks using Qwen2.5-3B (QLORA fine-tuned) with Faster-Whisper STT and local inference via Ollama, fully containerized with Docker Compose. Built a LangGraph-based reasoning agent that detects genuine ambiguity, asks a single targeted follow-up, and performs grounded state merging while preventing hallucinated fields (deadlines, people, or tasks). Integrated optional Notion execution API for real task creation, with deterministic agent traces and safe execution gating; system achieves >80% exact match on held-out evaluation after iterative dataset refinement.
View ProjectPak Econ Bot
June 24, 2026 – Present
Production agentic RAG system with a hand-rolled ReAct loop over LLaMA 3.3 70B; FastAPI backend on AWS EC2 and React frontend on AWS S3, deployed with automated CI/CD via GitHub Actions. Indexed ~1,200 chunks in Pinecone (384-dim cosine) with metadata filtering by section and content type; agent exposes 4 tools – semantic search, section lookup, table/numeric queries, and safe arithmetic evaluation. Switched production inference to ONNX Runtime instead of PyTorch, shrinking the Docker image by ~700 MB; per-session conversation memory with a 5-turn sliding window and isolated agent state per user.
View ProjectProject Inspector
June 24, 2026 – Present
AI-powered codebase analysis tool on Azure Container Apps with GitHub Actions CI/CD; parses Python AST across 15+ files in under 60 seconds. Renders dependency graphs via NetworkX and Graphviz; base64-embedded SVGs keep the API stateless and scalable. Dual-LLM fallback (Gemini 2.5 Flash → LLaMA 3.1 8B via Groq) with response cache cutting redundant calls by ~60%.
View ProjectCultural Fit Analysis
The candidate's diverse personal projects (Pak Econ Bot, Voice 2 Action, Project Inspector) showcase initiative, self-driven learning, and a passion for AI beyond academic or professional requirements. The experience in both LLM and computer vision domains, coupled with MLOps skills, indicates a broad technical curiosity and adaptability. The focus on production-readiness and measurable impact aligns with a results-oriented culture. The candidate's early career stage combined with advanced project work suggests a high potential for growth and a proactive attitude.
Soft Skills & Operational Fit
The candidate's project descriptions highlight a strong problem-solving aptitude, an ability to work with complex systems, and a focus on delivering measurable real-world impact. The detailed descriptions of architectural choices, optimizations, and evaluation metrics suggest a methodical and results-oriented approach. The use of CI/CD and containerization indicates an understanding of modern development and deployment practices, which aligns well with operational efficiency.