Open to opportunities

Sheraz Ahmad

@sherazahmad3

Message

I build production AI, Generative AI, RAG, and machine learning systems that improve performance and reduce costs.

Pakistan

Message

I've built and productionized AI systems at Prypco, Algo, Turing, and Afiniti, spanning Generative AI, RAG, forecasting, recommendations, fraud detection, and document intelligence. My work has reduced compute costs by 85%, cut processing time by 50%, and increased inference performance by up to 17%.

At Algo, I delivered LLM and RAG applications for Fortune 500 pharmaceutical clients under GDPR and HIPAA requirements, while migrating a flagship Forecasting Engine to Snowflake. At Turing and Afiniti, I improved real-time data and inference systems, including an Apache Spark indexing workflow that reduced runtimes by 80% and a LightGBM feature that improved business gain by 25%.

I'm also building multi-agent AI development platforms and secure multimodal legal-document intelligence products, bringing together LLM reasoning, cloud infrastructure, orchestration, and production deployment.

Experience

Work history, roles, and key accomplishments

Current

Lead AI Engineer

Current

Prypco

Jun 2025 - Present (1 year 3 months)

Improved production AI performance by owning model training, fine-tuning, deployment, and system optimization, increasing inference performance by up to 17%. Built a centralized feature store with Azure Data Lake, Delta Lake, and Redis, cutting model training time by 40%.

Model Training Fine Tuning Deployment System Optimization Azure Data Lake Delta Lake Redis AI Systems DevOps

Data Scientist, Product Engineering

Algo

Nov 2023 - May 2025 (1 year 6 months)

Built and released LLM and RAG applications for Fortune 500 pharmaceutical clients, combining vector retrieval, multimodal data processing, and Docker-based deployments while supporting GDPR and HIPAA requirements. Led migration of the flagship Forecasting Engine from Databricks Delta Lake to Snowflake, reducing monthly compute costs by 85% and cutting processing time by 50%.

LLM RAG Docker GDPR HIPAA Databricks Delta Lake Snowflake Terraform

Machine Learning Engineer

Turing

Sep 2022 - Oct 2023 (1 year 1 month)

Enabled real-time recommendation and decision systems by building streaming data pipelines with AWS Kinesis and Apache Flink. Replaced an Elasticsearch-based indexing workflow with Apache Spark on AWS EMR, reducing indexing runtimes by 80% and improving scalability for large data workloads.

AWS Kinesis Apache Flink Apache Spark Elasticsearch Data Pipelines Recommendation Systems Fraud Detection

Data Scientist II

Afiniti

Aug 2020 - Jul 2022 (1 year 11 months)

Improved feature extraction for production models by building NLP pipelines with Hugging Face, NLTK, and spaCy for sentiment analysis and text classification. Led the migration of a production framework from v5 to v6, completing validation and deployment one week ahead of schedule.

NLP Hugging Face NLTK SpaCy Sentiment Analysis Feature Engineering Pipeline Production Migration Frameworks

Data Scientist I

Afiniti

Mar 2019 - Aug 2020 (1 year 5 months)

Owned model development, feature engineering, and production pipelines for two major client accounts. Managed 5 production model units supporting $7.5M in monthly client revenue, contributing 6% to 12% incremental monthly sales.

Model Development Feature Engineering Pipeline Production LightGBM Client Management