Onur Barut, PhD

I build and evaluate production AI systems at scale — spanning multimodal models, retrieval, agents, inference infrastructure, typed data contracts, and evaluation. At Wizard AI, I lead product intelligence across a catalog of 70M+ products from heterogeneous retail feeds. The platform combines text and images with hybrid BM25 and vector retrieval, schema-constrained generation, and deterministic verification to improve categorization, attribute enrichment, entity mapping, and catalog quality. I focus on making model quality measurable and systems dependable: frozen evaluation sets, regression gates, cost and latency comparisons, and reconciliation checks that catch data lost between pipeline stages. Before Wizard I was an AI Software Architect at Intel, working on multimodal RAG, VLM-generated synthetic data, and hardware-aware inference. I earned my PhD in Computer Engineering at UMass Lowell and developed a graduate generative AI course for Clark University. I am open to Staff or Lead roles in applied AI, GenAI, and search and recommendations.

A snapshot of what I’m building and sharing, updated September 29, 2026.

• Building

  FlipMap — a web product built with Next.js and server-sent events.

• Research

  Market-structure tooling — structural-arbitrage scanning and a five-pillar premarket scanner.

• Exploring

  AI Chemist LLC — vending pilots and hands-on product experimentation.

• Sharing

  Graduate GenAI notebooks — public exercises from the course I developed for Clark University.

Beyond production ML: independent products, quantitative tooling and interactive experiments.

• FlipMap

  A web product built with Next.js and server-sent events, extending my work on streaming systems into an independent product.

  Product engineering · Next.js · SSE

• Kalshi structural-arbitrage scanner

  Quantitative tooling for investigating relationships between prediction markets and identifying structural pricing discrepancies.

  Market-structure research · Prediction markets

• Five-pillar premarket scanner

  A premarket research scanner that organizes signals into a five-pillar framework for review.

  Quantitative tooling · Research workflow

• AI Chemist LLC

  An entrepreneurial venture exploring vending pilots, taking product experimentation beyond software.

  Entrepreneurship · Vending pilots

• Wizardic Quest

  A playable browser game with exploration, combat, inventory and daily quests. A playground for rendering and game-loop experiments.

  HTML5 Canvas · ES modules · Playable prototype

Platform, model and data-quality work for product intelligence across a catalog of 70M+ products. Company-internal names, customers and systems are deliberately omitted.

• AI catalog enrichment platform

  51 registered job types

  Owner and primary author of successive generations of the service that turns heterogeneous retailer feeds into search-ready categories, typed attributes, summaries and embeddings. The distributed pipeline combines multimodal inputs, schema-constrained generation, batch inference and deterministic validation. It has processed roughly 3–4 million product enrichments in about two weeks during large runs, and now runs whole-catalog campaigns unattended with continuous downstream ingestion.

  • Python
  • FastAPI
  • Celery
  • Pydantic v2
  • Spark
  • Airflow
  • Parquet
  • GCS
  • Cloud Run

• Recategorization at catalog scale

  34.0% → 44.9% on a 256-product audit set

  Owned LLM product categorization into a ~6,100-leaf taxonomy across three engine generations, ending hallucinated category paths by generating strict Literal-typed response schemas over retrieved candidates. Replayed an operations error set into a repeatable evaluation loop: on a separate 138-product operations error set, exact-category fixes rose 20 → 39 and right-department fixes 33 → 78 with zero regressions. A controlled A/B on 1,199 random listings with retrieval held fixed lifted agreement with production labels 90.2% → 91.0% (19 wins, 9 losses).

  • Structured outputs
  • Hybrid retrieval
  • Elasticsearch
  • vLLM
  • OpenAI Batch API

• Multimodal product recategorization

  ~79% → ~91% on a curated evaluation set

  Combined product text, images, metadata, embeddings and hybrid retrieval to improve category assignment. Candidate retrieval narrows the decision space; schema-constrained model output and deterministic checks keep results grounded and valid.

  • Vision-language models
  • Elasticsearch BM25 + kNN
  • RRF
  • Typed outputs

• Catalog audit agent

  359 labeled cases · 18 agent-only wins vs 5 workflow-only

  Designed an agent for evidence gathering, with deterministic code retaining input validation, evidence grounding, verification, policy enforcement and routing. Compared the agent and workflow baseline using the same tools and policies: paired correctness favored the agent (p ≈ 0.011), and on 110 clean products it produced 5 false flags versus 15 for the workflow.

  • Agent evaluation
  • Tool use
  • Paired analysis
  • False-positive measurement

• Typed contracts and data integrity

  Found a schema mismatch dropping ~81% of batch values

  Connected Pydantic, Spark and Parquet stages with explicit schema checks, row and value reconciliation, serialization regression coverage, and quarantine paths. The incident showed why job completion alone is not proof that intended information survived the pipeline.

  • Pydantic
  • Spark
  • Parquet
  • Data contracts
  • Reconciliation

• Retailer feed mapping service

  87–99% per-attribute accuracy (median ~93%), 11 garment-size attributes

  Built and owned a Cloud Run service that onboards retailer feeds onto a ~6,750-category taxonomy and a ~1,450-attribute schema: hybrid Elasticsearch retrieval (BM25 + kNN + RRF) proposes candidates and an LLM selects under a dynamically generated Literal schema, so it cannot invent a label. Also cross-mapped a next-generation attribute schema onto the legacy repository, covering roughly 1,280 of ~1,450 attributes (about 88%).

  • Elasticsearch
  • BM25 + kNN + RRF
  • GraphQL
  • Cloud Run
  • Docker

• Orchestration and release ownership

  2× batches in flight (50 → 100)

  Took over the enrichment DAG family in the data-engineering team's Airflow repository across roughly 100 pull requests. Improved filters and per-backend concurrency, and built an idempotent Cloud Run operator that persists the operation name and re-attaches to an in-flight job on retry. Diagnosed that DAG limits and pools ignore deferred tasks — the limits looked correct but bounded nothing — and shipped per-backend pools.

  • Apache Airflow
  • Cloud Composer
  • Cloud Run
  • PySpark
  • pytest

• Open-weight models, chosen on evidence

  Self-hosted 35B edged GPT-4.1-nano, 0.714 vs 0.698

  Led the move of catalog enrichment from per-token vendor APIs to self-hosted open-weight models (Qwen3 / 3.5 / VL, GPT-OSS, Gemma on vLLM behind a LiteLLM gateway), landing a 68-file, ~20.7k-line integration, keeping the vendor batch path as a pluggable fallback, and holding production up through a compute-capacity interruption by re-pointing a live job to an external GPU-cloud model on a two-line config change. Models were picked as an evaluation problem, not a demo: frozen and held-out sets, per-category slices, severity-weighted regression gates, structured-output and evidence checks, judge calibration, and latency and cost comparisons. A calibrated text-and-vision judge plus reference-free rubrics showed an 8B open-weight model matching an 80B on 754-variant name normalization and an open-weight VLM producing numeric hallucinations about 3× as often (4.4% vs 1.4%), prompting additional numeric safeguards.

  • vLLM
  • LiteLLM
  • Qwen3 / Qwen3-VL
  • GPT-OSS
  • LLM-as-judge

• Catalog data-quality program

  Applied size coverage 11% → 79% on one feed cohort

  Turned anecdotal quality complaints into measured programs: read-only Elasticsearch audits quantified ~4.7M documents that carried a valid size prefix but no size fields, and a normalized size attribute out of enum on ~31% of ~11.2M documents. The fix was deterministic rather than another LLM pass — a fill-only, evidence-ledgered, dry-run-by-default pipeline routed through the standard ingestion path. Separately built a conformance validator that checked 26 enrichment configs (25 per-vertical plus a fallback) against the schema and taxonomy and enumerated 1,044 conformance issues across eight classes — 365 of them low-severity name-similarity pairs — against staging snapshots, before any of them reached production.

  • Elasticsearch aggregations
  • Delta Lake
  • Dry-run/ledger design
  • Airflow
  • pandas

• Near-duplicate image detection

  Rerank-latency benchmark

  A research benchmark for near-duplicate listing detection in a recommendation rerank step under a hard latency budget: a classical hash / HSV / ORB + RANSAC cascade against triplet-loss fine-tuned CNN backbones and a multimodal embedding model with Matryoshka truncation, all through one harness that sweeps thresholds for precision/recall and measures warm and cold latency against explicit PASS/FAIL budgets.

  • PyTorch
  • OpenCV
  • Triplet loss
  • Qwen3-VL-Embedding
  • Streamlit

• Conversational shopping features

  Catalog-grounded

  Worked on the shopping assistant's prompt layer — soft-constraint entity extraction, brand and entity extraction merged into one async call, and hard-coded size lists replaced with values fetched from the catalog API — then an LLM-judged pipeline that decides which attributes a category genuinely requires, so the assistant asks only the clarifying questions that category actually needs. Later prototyped a retrieval-grounded, steerable shopping experience over a hybrid BM25 + kNN pipeline.

  • Prompt engineering
  • FastAPI
  • SSE
  • Elasticsearch RRF
  • MongoDB

• Multimodal RAG and video super-resolution

  Open-sourced at Intel

  As AI Software Architect at Intel: trained and open-sourced a multimodal visual RAG model on a LLaMA backbone, generated VLM-based synthetic data to fine-tune a smaller language model for document question answering, and open-sourced a video super-resolution model reported at +2% PSNR with reduced artifacts at minimal additional compute.

  • LLaMA
  • Multimodal RAG
  • PyTorch
  • OpenVINO
  • ONNX

1. Wizard AI — Lead Machine Learning Engineer

   Oct 2024 – Present · Remote · MA

   AI shopping company. Owner of the AI catalog-enrichment platform: categories, typed attributes, summaries and embeddings for a search index of 70M+ products from heterogeneous retailer feeds.

   • Owned the AI catalog-enrichment platform across two service generations and three repositories, taking it from a research-grade service with manual file handoffs to a job framework of 51 registered jobs that runs whole-catalog campaigns unattended and is ingested continuously by the data platform.
   • Led the migration of catalog enrichment from per-token vendor APIs to self-hosted open-weight models (Qwen3 / 3.5 / VL, GPT-OSS, Gemma on vLLM behind a LiteLLM gateway), selecting each model on benchmark evidence — a self-hosted 35B model edged GPT-4.1-nano on a 381-variant summary benchmark (0.714 vs 0.698) — keeping the vendor batch API as a pluggable fallback, and holding production up through a compute-capacity interruption by re-pointing a live job to an external GPU-cloud model on a two-line config change.
   • Selected models per task with evaluation-first benchmarks: 754-variant name normalization (98.9% accuracy for the best vendor-API model, with an 8B open-weight model matching the 80B), 1,077-variant vision-language recategorization (92.8% full-path exact match for the leading vendor model; the open-weight 8B led subcategory accuracy), and a 381-variant summary benchmark.
   • Owned LLM product categorization into a ~6,100-leaf taxonomy across three engine generations, eliminating hallucinated category paths by generating strict Literal-typed response schemas over retrieved candidates.
   • Ran catalog-scale campaigns: ~1.7M listings in a taxonomy migration, a multi-million-listing recategorization campaign, 9M+ records lite-enriched, ~3.5M listings size-enriched across 100+ retailer feeds, and a ~275k-product brand pass across 100+ retailer feeds.
   • Migrated catalog embeddings from a paid vendor API to a self-hosted fine-tuned Qwen3 model, backfilled 10 product categories, and fixed a silent data-loss bug so the full set of per-facet vectors reaches the search index.
   • Took over the enrichment Airflow DAG family in the data-engineering team's repository (roughly 100 pull requests), diagnosing that DAG limits and pools ignore deferred tasks and shipping per-backend pools that doubled batches in flight (max active runs 50 → 100), plus an idempotent Cloud Run operator that re-attaches to an in-flight job on retry.
   • Shipped a deterministic, LLM-free size and color fix pipeline — fill-only, evidence-ledgered, dry-run by default, routed through the standard ingestion path — that lifted applied size coverage 11% → 79% on one feed cohort — 100% LLM-judged accuracy on the variant-evidenced letter/numeric/waist tiers (313/313) and 98% on the text-derived default tier — validated on a staging cohort ahead of production rollout.

2. Clark University — Adjunct Faculty (part-time)

   Fall 2025 · Clark University

   • Developed MSIT 3103: Generative AI Model Development, a graduate course on building with generative AI models, for Fall 2025.
   • Published the course assignment notebooks as a public repository so students — and anyone else — can run the exercises: MSIT3103-GenAI-assignments.

3. Intel Corporation — AI Software Architect

   Jun 2022 – Oct 2024 · Remote / MA

   • Generated synthetic data with a vision-language model to fine-tune a smaller language model for document question answering.
   • Built a video-understanding proof of concept for retail video analytics: created and annotated a task-specific dataset, fine-tuned VideoLLaMA, generated AdaCLIP video embeddings, indexed them in VDMS, and retrieved relevant segments for natural-language queries. Also trained and open-sourced a multimodal visual RAG model on a LLaMA backbone.
   • Designed and open-sourced deep learning models for network-AI projects and optimized them for Intel hardware against customer compute budgets, reported at up to 50% higher throughput.
   • Developed and open-sourced a video super-resolution model reported at +2% PSNR with reduced artifacts at minimal additional compute cost.
   • Led research on network security and video quality enhancement, publishing and presenting more than five papers internally and externally.

4. University of Massachusetts Lowell — Research Assistant, ACANETS Lab

   Aug 2018 – May 2022 · MA

   • Raised malicious traffic detection above 90% and improved model expansion capability with a transformer-based deep learning architecture.
   • Co-organized the NetML 2020 network traffic analytics competition, and collected and open-sourced the HAR-UML20 human activity recognition dataset.
   • Developed a multi-task LSTM model for activity classification and intensity estimation with over 95% accuracy.

5. Intel Corporation — Graduate Intern for Machine Learning

   May – Aug 2019 · CA

   • Built end-to-end machine learning pipelines for network traffic analysis and improved throughput by 15% by optimizing the flow feature selection algorithm.

6. Aselsan — System Design Engineer

   Feb 2014 – Jul 2018 · Turkey

   • Led the team responsible for flight-test data acquisition tasks and system integration tests.
   • Increased flight-test data acquisition speed 5× by replacing the PCM interface with Ethernet.
   • Contributed to Electronic Warfare Self-Protection (EWSP) integration projects with flight data acquisition.

Ordered by depth — the first items in each group are where I have designed and owned production systems.

★A star marks the three most-cited papers; citation counts shown only for those.

1. RET-Net: A CNN Framework for Real-Time Traffic Classification Using Key-Byte Mechanism

   IEEE Transactions on Network and Service Management, 2026.

2. Malware Detection for Portable Executables Using a Multi-input Transformer-Based Approach

   IEEE ICNC, 2024. DOI — Malware Detection for Portable Executables Using a Multi-input Transformer-Based Approach

3. AWS Compute Video Super-Resolution powered by the Intel® Library for Video Super Resolution

   Mile-High Video Conference (MHV '24, ACM), 2024. DOI — AWS Compute Video Super-Resolution powered by the Intel® Library for Video Super Resolution

4. Universal Network Traffic Analysis for Malicious Traffic Detection using RappNet: A Privacy-Preserving Approach

   IEEE CNS 2023 (poster). DOI — Universal Network Traffic Analysis for Malicious Traffic Detection using RappNet: A Privacy-Preserving Approach

5. A Comprehensive Study on Efficient and Accurate Machine Learning-Based Malicious PE Detection

   IEEE CCNC, 2023. DOI — A Comprehensive Study on Efficient and Accurate Machine Learning-Based Malicious PE Detection

6. Weakly supervised learning for network traffic classification

   IEEE NAS, 2022. DOI — Weakly supervised learning for network traffic classification

7. ★Most cited. R1DIT: Privacy-Preserving Malware Traffic Classification with Attention-Based Neural Networks

   IEEE Trans. Network and Service Management, 2022 (early access). · 69 citations. DOI — ★Most cited. R1DIT: Privacy-Preserving Malware Traffic Classification with Attention-Based Neural Networks

8. Multi-Task Hierarchical Learning Based Network Traffic Analytics

   IEEE ICC, 2021. DOI — Multi-Task Hierarchical Learning Based Network Traffic Analytics

9. Machine learning based malware detection on encrypted traffic: A comprehensive performance study

   NSysS 2020 (ACM), 2020. DOI — Machine learning based malware detection on encrypted traffic: A comprehensive performance study

10. TLS Encrypted Application Classification Using Machine Learning with Flow Feature Engineering

    ICCNS 2020 (ACM), 2020. DOI — TLS Encrypted Application Classification Using Machine Learning with Flow Feature Engineering

11. ★Most cited. Multitask LSTM Model for Human Activity Recognition and Intensity Estimation Using Wearable Sensor Data

    IEEE Internet of Things Journal, 2020. · 95 citations. DOI — ★Most cited. Multitask LSTM Model for Human Activity Recognition and Intensity Estimation Using Wearable Sensor Data

12. ★Most cited. NetML: A Challenge for Network Traffic Analytics

    2020. · 50 citations. arXiv

13. Geospatial object detection using deep networks

    SPIE Proceedings, 2019. DOI — Geospatial object detection using deep networks

Full list and citation counts on Google Scholar.

Datasets, challenge baselines, paper code and course material.

• NetML Competition 2020

  Network traffic analytics challenge I co-organized: open dataset, baselines and evaluation for malware and traffic classification.

  26 stars · co-organizer

• HAR-UML20

  Human activity recognition dataset I collected and open-sourced with the ACANETS Lab.

  2 stars · dataset

• Encrypted Malware Detection

  Code for the comprehensive performance study of machine learning malware detection on encrypted traffic.

  6 stars · paper code

• MultiTask-LSTM-HAR

  Code for the IEEE Internet of Things Journal multi-task LSTM activity recognition paper.

  3 stars · paper code

• Multi-Task Hierarchical Learning

  Code for the multi-task hierarchical network traffic analytics paper.

  2 stars · paper code

• Encrypted App Classification

  Code for the TLS encrypted application classification paper with flow feature engineering.

  paper code

• MSIT3103 GenAI Assignments

  Assignment notebooks for the graduate generative AI course I developed at Clark University.

  course material

I was part-time faculty at Clark University, where I developed MSIT 3103: Generative AI Model Development, a graduate course for Fall 2025. The assignment notebooks are public, so the exercises can be run by anyone: MSIT3103-GenAI-assignments. Earlier, as a graduate student at UMass Lowell, I was a Teaching Assistant for the Logic Design course in Fall 2018, coordinating lab work and running tutoring sessions.

I am open to new opportunities and always happy to talk about production LLM systems, search and retrieval, or evaluation — reach me at any of these.

• onurbarut@windowslive.com
• LinkedIn
• GitHub
• Google Scholar