Nabeel Gulzar.

Hello, I'm

Nabeel Gulzar

I build LLM systems|

I architect scalable, end-to-end NLP and LLM-driven systems: evaluation platforms, auto-reviewers and agentic workflows that turn research into measurable business impact.

See my workGet in touch

About

Team lead who ships: from LLM auto-reviewers to async QC platforms on GCP.

Fine-tuning (SFT, RLHF), prompt engineering and rigorous evaluation.

Roots in NLP research, search relevance and full-stack data visualisation.

Measurable impact

35%

Reduction in average handle time

94%

Precision/recall on LLM auto-reviewers

5–6%

Search conversion uplift

96%

CI/CD test coverage on Auto QC

Experience

  1. AI Engineer & Team Lead

    Turing Inc. · Remote, UK

    04/2024 – 05/2026

    Progressed from Data Scientist to Pod Lead to Team Lead, owning model reasoning frameworks, SFT/RLHF tuning and automated QC pipelines.

    • Team Lead – AI: built AI infrastructure, auto-reviewer systems and high-throughput evaluation platforms.
    • Reduced average handle time by 35% by automating data-curation scripts.
    • Optimised auto-reviewer LLMs (ambiguity, realism, path correctness) to 94% precision/recall with self-feedback loops.
    • Designed Auto QC Service on GCP (FastAPI, MongoDB, Redis, Cloud Tasks, Docker) with 96% CI/CD test coverage.
    • Built LLM evaluation frameworks to balance prompt quality, model choice and latency/cost across multi-step chains.
    LLMsSFTRLHFGCPFastAPIRedisMongoDB

  2. Senior Data Scientist

    Daraz (Alibaba Group) · Remote, PK

    08/2022 – 08/2023

    Expanded product knowledge graphs for search and recommendations.

    • Achieved a 5–6% conversion uplift via query rewrites and taxonomy mapping from user-behaviour logs.
    • Built NLP pipelines extracting structured attributes from reviews and query logs.
    • Deployed query-category assignment to filter non-related products from search.
    • Designed A/B tests to measure algorithmic impact on core business KPIs.
    NLPPyTorchHugging FaceKnowledge GraphsA/B Testing

  3. Research Assistant

    LUMS · PK

    06/2021 – 07/2022

    Research in NLP alongside an MS in Computer Science.

    • Implemented state-of-the-art NLP models.
    • Co-authored: A Comparative Analysis of Clone Detection Techniques on SemanticCloneBench (2022).
    NLPResearch

  4. Platform Engineer

    Intech Process Automation · PK

    06/2019 – 04/2021

    Full-stack work on industrial data visualisation.

    • Built interactive data charts (pie, multiline, Pareto, stacked bar) in React and D3.
    • Implemented backend transport for large volumes of sensor data.
    • Established the testing framework, improving coverage and reliability.
    ReactD3Backend

  5. Data Analyst

    Techlogix · PK

    10/2018 – 06/2019

    Where the data journey started.

    • Managed ETL pipelines for enterprise clients using MS SSIS.
    • Built Power BI dashboards with row-level security.
    ETLSSISPower BI

Skills

Tap a category to highlight it.

AI & Architecture

LLMsAgentic FrameworksSFT / RLHFNLPDeep LearningPyTorchTensorFlowHugging Face

Backend & Data

PythonFastAPIFlaskDjangoProtobufSQLMongoDBRedisPandasNumPy

Cloud & MLOps

GCPAWSCloud RunCloud TasksDockerGitHub ActionsLinuxA/B Testing

Languages

PythonSQLJavaScriptC++

Education

MS Computer Science

LUMS · 2020 – 2022

BS Computer Science

NUCES · 2014 – 2018

Bronze medal · Dean's List

Let's build something together

Open to conversations about LLM systems, evaluation, and AI infrastructure.

nabeelahmadgulzar@gmail.comLinkedIn ↗GitHub ↗

© 2026 Nabeel Gulzar · Berlin, Germany