Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Rafe Tariq, an AI engineer specializing in production agentic systems, RAG, and LLM infrastructure. I've shipped LLM platforms at three VC-backed startups, including as Founding CTO of a ConceptionX-backed company, and I'm based in London, UK. I earned a PhD in Artificial Intelligence from the University of Manchester and an MEng in Computer Science from the University of York. My work spans building scalable data pipelines, evaluating prompts and retrieval, and delivering production-grade AI for enterprise contexts.…I'm Rafe Tariq, an AI engineer specializing in production agentic systems, RAG, and LLM infrastructure. I've shipped LLM platforms at three VC-backed startups, including as Founding CTO of a ConceptionX-backed company, and I'm based in London, UK. I earned a PhD in Artificial Intelligence from the University of Manchester and an MEng in Computer Science from the University of York. My work spans building scalable data pipelines, evaluating prompts and retrieval, and delivering production-grade AI for enterprise contexts.

Rafe Tariq

AI Engineer, Data Scientist, Full Stack Developer, +4





I'm Rafe Tariq, an AI engineer specializing in production agentic systems, RAG, and LLM infrastructure. I've shipped LLM platforms at three VC-backed startups, including as Founding CTO of a ConceptionX-backed company, and I'm based in London, UK. I earned a PhD in Artificial Intelligence from the University of Manchester and an MEng in Computer Science from the University of York. My work spans building scalable data pipelines, evaluating prompts and retrieval, and delivering production-grade AI for enterprise contexts.…I'm Rafe Tariq, an AI engineer specializing in production agentic systems, RAG, and LLM infrastructure. I've shipped LLM platforms at three VC-backed startups, including as Founding CTO of a ConceptionX-backed company, and I'm based in London, UK. I earned a PhD in Artificial Intelligence from the University of Manchester and an MEng in Computer Science from the University of York. My work spans building scalable data pipelines, evaluating prompts and retrieval, and delivering production-grade AI for enterprise contexts.

Available to hire

I’m Rafe Tariq, an AI engineer specializing in production agentic systems, RAG, and LLM infrastructure. I’ve shipped LLM platforms at three VC-backed startups, including as Founding CTO of a ConceptionX-backed company, and I’m based in London, UK.

I earned a PhD in Artificial Intelligence from the University of Manchester and an MEng in Computer Science from the University of York. My work spans building scalable data pipelines, evaluating prompts and retrieval, and delivering production-grade AI for enterprise contexts.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

English

Fluent

Work Experience

AI Engineer (Contract) at Intuigence AI

August 1, 2025 - October 1, 2025

Shipped production RAG and agentic workflows for enterprise document processing; cut retrieval latency ~30% vs. baseline. Built multi-modal extraction pipeline on Llama 4 Maverick combining OCR, layout parsing, and structured output for regulated docs. Instrumented stack end-to-end with OpenTelemetry + Sentry so failures and model regressions surface within minutes. Built LLM eval harness (LangSmith + custom regression tests) so prompt and model changes ship without silent accuracy drops.

A.I. Engineer (Contract)

August 1, 2025 - October 1, 2025

Developed AI-powered multi-modal document processing and intelligent search/retrieval. Built RAG pipelines and agentic workflows, improving document retrieval efficiency by 50%. Implemented microservices architecture with real-time WebRTC and hybrid search, enhancing platform scalability.

Founding CTO at Paynto AI

January 1, 2024 - July 1, 2025

Founded and led engineering for an agentic AI platform automating genomic/proteomic analysis; +35% research throughput for early customers. Designed scalable ingest and processing pipelines for large biological datasets, cutting time-to-insight ~40%. Set technical direction end-to-end: model selection, infra (AWS/Kubernetes), hiring, and customer deployments. Built evaluation tooling over domain tasks so agent updates could be benchmarked on accuracy, latency, and cost before merging.

CTO / A.I. Engineer at Pay to A.I.

January 1, 2024 - July 1, 2025

Led development of an agentic AI platform for biosciences, enabling automated analysis of genomic and proteomic data, improving research throughput by 35%. Designed multi-agent system architecture with RAG and knowledge graphs, enhancing data-driven insights for drug discovery and biomarker identification. Built scalable data pipelines for processing large-scale biological datasets, reducing processing time by 40%.

EIR (CTO track) at Entrepreneur First

April 1, 2023 - August 1, 2023

Co-founded a GenAI co-pilot for enterprise automation; onboarded 3 pilot customers during the cohort. Engineered RAG over internal knowledge bases, reducing retrieval latency ~65% in high-volume workflows. Led full-stack MVP delivery (intelligent scraping, orchestration, frontend) two months ahead of target. Closed the loop from pilot feedback into prompt + retrieval evals, so each iteration measurably improved task success rate.

A.I. Engineer at Entrepreneur First

April 1, 2023 - August 1, 2023

Founded GenAI co-pilot startup, building enterprise-grade automation tools adopted by three pilot customers. Engineered RAG pipelines tailored to internal knowledge bases, cutting information retrieval latency by 65% in high-volume user workflows. Led full-stack product development including intelligent scraping and front-end/back-end orchestration, accelerating MVP delivery by two months.

AI Engineer (part-time) at Singularity Labs Fund

February 1, 2022 - May 1, 2024

Built and operated data infrastructure for high-throughput quant research; cut experimentation cycle time ~30%. Optimised petabyte-scale Spark + Kafka pipelines; improved ingest latency and downstream accessibility ~20%. Productionised Transformer-based models on Cerebras CS for research backtests. Prototyped long-context sequence-model fine-tuning on Cerebras CS; lower training wall-time vs. GPU baseline for extended-context runs.

A.I. Engineer at Singularity Labs Fund

February 1, 2022 - May 1, 2024

Scaled transformer models (Spacetimeformer) on financial market and social sentiment data, enhancing predictive accuracy for trading signals by 18%. Designed and deployed data infrastructure supporting high-throughput research workflows, reducing model experimentation time by 30%. Built and optimized petabyte-scale data pipelines with Spark and Kafka, improving data ingestion latency and accessibility by 20%.

AI Researcher (External) at Imperial College London

July 1, 2021 - September 1, 2022

Wildfire prediction with CNNs + autoencoders over topographic data; 25% higher accuracy vs. baseline heuristics. Integrated GANs, UNets, VAEs into forecasting pipelines — higher-resolution spatial predictions, 22% fewer false positives. Benchmarked against ResNet and VGG-16; matched accuracy with ~40% fewer parameters. Released reproducible training + evaluation code; reused by subsequent student projects extending the wildfire work.

A.I. Researcher (External) at Imperial College London

July 1, 2021 - September 1, 2022

Developed wildfire prediction models using CNNs and Autoencoders on topographic data, achieving 25% higher accuracy vs. baseline heuristics. Integrated GANs, UNETs, and VAEs into forecasting pipelines, increasing resolution of spatial predictions and reducing false positives by 22%. Benchmarked models against pretrained networks (ResNet, VGG-16), demonstrating comparable performance with 40% fewer parameters.