Engineer, builder, researcher in training.
I work on data systems for text-heavy domains: ingestion, structure, retrieval, evaluation, and the operational discipline needed to keep those systems trustworthy.
I am a data engineer based in Lahore, currently leading the data layer at DigiLawyer. My work sits between backend data infrastructure and applied information retrieval: turning fragmented legal sources into structured, searchable, and AI-usable records.
The common thread across my projects is reliability under messy real-world constraints: scanned PDFs, inconsistent metadata, changing source formats, production deadlines, and users who need answers they can verify.
Degree
BSc Computer Science, University of Engineering and Technology Lahore
Focus
Data engineering, information retrieval, NLP systems, databases, and cloud infrastructure
Final-year project
MyContract: AI-assisted contract generation, review, and compliance checking; later adapted into DigiLawyer's Mike AI Draft feature.
Questions I want to study further.
- ›Hybrid legal retrieval: lexical, dense-vector, and re-ranking systems for statutes and judgments.
- ›Temporal legal data models: representing amendments, versions, and section-level provenance.
- ›Reliable data infrastructure for public-interest corpora where correctness and traceability matter.
- ›Evaluation methods for retrieval-augmented systems: recall, citation coverage, latency, and regression tests.
What I can contribute now.
- ›Production data systems: ingestion, validation, indexing, and operational recovery.
- ›Search and retrieval: PostgreSQL FTS, pgvector, MeiliSearch, Qdrant, hybrid scoring, and corpus design.
- ›Applied AI engineering: LLM-backed workflows where outputs need citations, structure, and reviewability.
- ›Leadership in small teams: documentation, handoff discipline, intern supervision, and data quality ownership.