Faraz Ali

AI Product Engineer (LLM Focused) — Invictus Core

I am an AI Product Engineer with 3+ years of experience building and shipping production-grade AI systems for international teams and frontier AI teams. I specialise in large language models, Voice AI, RAG pipelines, and agentic workflows — combining strong ML fundamentals with hands-on product execution. I am open to senior remote AI engineering roles, and also available for select contract work.

Since June 2026 I have led the ML engineering for a stealth customer-intelligence product built for a billion-dollar US apparel brand (anonymized under NDA), which made me fluent in data at scale: Azure Data Lake and Azure ML jobs, AWS Glue and Athena, DuckDB and Parquet. The numbers are on the case study page.

I work directly with teams to own the full product lifecycle: from problem definition and technical decisions through to implementation and deployment. I also run free mentorship sessions for students breaking into AI (see invictus-core.dev/mentorship).

What I Do

• LLM Integration & Product Engineering — End-to-end integration of OpenAI, Anthropic, and open-source models into production systems. Model selection, prompt design, workflow architecture, evaluation, and deployment.
• RAG Pipeline Design — Retrieval-Augmented Generation systems using vector databases, ElasticSearch, and embedding strategies for grounded AI responses.
• Voice AI Systems — Automated voice interaction pipelines including transcription, intent detection, and LLM-driven speech responses. Built a live Voice AI system for healthcare hospitals.
• Agentic Workflows — Autonomous AI agents that execute multi-step tasks, integrating LLM reasoning with real-world tool use and APIs.
• Generative Engine Optimization (GEO) — Pipelines to improve brand and product visibility across LLM-based search and answer engines through structured content, embeddings, and retrieval strategies.
• AI Product Prototyping — Rapidly building and shipping MVPs and production-ready AI applications, often owning both backend and frontend.

Selected Projects

• Customer Intelligence Platform (stealth, anonymized) — Scout personas over 57M customer profiles; Offer Engine deciding in 0.08 ms with margin protection; 2.86B-row clickstream reduced from 1 TB to 7 GB; 17.6M anonymous visitors scored. Case study.
• Voice AI for Healthcare Hospitals — Automated patient interactions with transcription, intent detection, and LLM-driven responses with domain-specific reliability constraints.
• Autonomous SDR Platform — Full-ownership AI platform for automated sales outreach with LLM-powered personalisation at scale.
• E-commerce Product Mapping Platform — LLM-powered catalog analysis system for structured product representations, discovery, and automation.
• Business Development App — LinkedIn network analysis using LLMs to cluster connections, identify ICPs, and recommend engagement strategies.
• Co-Working Meeting Bot — Real-time meeting transcription, summaries, action items, and agenda suggestions integrated into a virtual office platform.
• Unified LLM Integration Layer — Abstraction layer across LLM providers with centralised authentication and standard payload schema, reducing maintenance overhead by 40%.

Tech Stack

• Python
• DuckDB
• Parquet
• AWS Athena
• AWS Glue
• Azure Data Lake
• Azure ML
• ClickHouse
• TypeScript
• FastAPI
• LangChain
• LlamaIndex
• OpenAI API
• Anthropic API
• RAG
• ElasticSearch
• Redis
• PostgreSQL
• MongoDB
• Neo4j
• WebSockets
• asyncio
• Docker
• Kubernetes
• PyTorch
• TensorFlow
• AWS

Experience

• LLM Product Engineer — D-Cube Tech (Dec 2025 – Present, Remote)
  Building and shipping AI products for international clients, owning the full lifecycle from problem definition to deployment. Since June 2026: ML lead on a stealth customer-intelligence product for a billion-dollar US apparel brand.
• ML Engineer (LLM Focus) — Turing (Nov 2025 – Jan 2026, Remote, US)
  Collaborating with frontier AI teams including Anthropic and Meta on LLM performance, reasoning, alignment, and inference-time efficiency.
• Lead AI Product Engineer — QLU.ai (May 2024 – Nov 2025, Karachi)
  Led the AI engineering team across multiple AI-driven products with full technical and product ownership.
• Data Science Research Assistant (Dec 2023 – Nov 2024, London, UK)
  Ph.D. research collaboration on NLP sentiment analysis across 172 UK public companies.

Education

• B.S. Computer Science — Habib University (2020–2024)
  CGPA: 3.75/4.00 · Concentration: Deep Learning & Data Science
  Thesis: LipSync Voice in Urdu for Aphonia Patients

Availability

Open to senior remote AI engineering roles, with a preference for async-friendly teams. Also available for select contract work involving LLM integration, RAG systems, Voice AI, or agentic product development.