
Hasnain Ali
Senior AI Engineer
6+ years shipping software, 5+ in AI, starting with RLHF training data at Turing. I build AI voice agents and document AI, and the Python backends they run on.
Remote from Pakistan (UTC+5). Open to relocation. US roles need visa sponsorship.
Career
Experience
Jul 2026 – now
Independent contractor, remoteFreelance AI Engineer
Building Mentanima, a legal copilot that cites its source for every answer, and Automation Studio, four agent workflow demos you can run in your browser.
Nov 2024 – Jun 2026
Full-timeSenior Software Engineer at Devsinc
Led eight AI engineers across legal, healthcare and real-estate products, including retrieval across about five million documents.
Senior AI Lead Engineer on Smart Advocate: led its document AI, from OCR, classification and structured extraction to knowledge-graph retrieval, with a 40% accuracy gain over standard RAG.
Mar 2022 – Nov 2024
Full-timeAssociate Software Engineer at Amrood Labs
Django APIs on PostgreSQL and AWS for fleet and employee management software.
Contract analysis on the OpenAI API that cut manual review by about 60%.
Earlier
Jan 2022 – Aug 2023
Part-timeDjango Developer at Arbisoft
Apr 2021 – Feb 2022
Full-time, remoteSoftware Engineer, POD lead at Turing
Led an annotation team producing RLHF training data for a frontier AI lab.
Mar 2020 – Mar 2021
Founder and lead engineer at 2ndPlace
Built a Django and React travel platform: 3,500+ active users, 15,000+ listings.
Skills
Agents: , , , , , ,
Models and tuning: , , ,
Retrieval: , , ,
Backend: , , , , ,
Selected systems
Speculo.ai
Senior AI Communication Engineer
AI that calls, texts and emails real-estate leads, then hands the qualified ones to the client's human agents. It contributed to $2M+ in annual revenue.
My part: the Python backend that times every call, text and follow-up.
117K+ emails and SMS, plus about 10K AI voice calls, in one 6-hour run.
How the pipeline worksPython, LangGraph, OpenAI, Twilio, ElevenLabs, GCP Pub/Sub, pgvector
Smart Advocate
Senior AI Lead Engineer, at Devsinc
Document AI for legal and medical case files, shipped into a platform used by 20,000+ legal professionals.
I led the AI work: OCR, classification and extraction, and the knowledge-graph retrieval.
40% accuracy gain over standard RAG.
Knowledge graph plus a fine-tuned model. Answers scored by a GPT-4-class judge and BLEU/ROUGE, retrieval by precision@k and recall@k.
How it was measuredPython, LangGraph, RAG, KAG, Neo4j, LoRA
Mentanima
In development
A legal copilot: document search backed by a knowledge graph. Every answer cites its source, or says it can't answer.
Python, RAG, KAG, Neo4j, LangGraph
Automation Studio
Four agent workflow demos you can run in your browser.
Try them ↗Python
toon-mcp-server
An open-source MCP server on PyPI. It converts JSON and system prompts to TOON, a compact format that uses fewer tokens.
PyPI ↗Python, MCP
Education
BS Computer Science, COMSATS University Islamabad
Sources
- Speculo.ai: hasnainali.vercel.app/blog/gcp-pubsub-ai-outreach
- Smart Advocate: hasnainali.vercel.app/blog/kag-vs-rag
- Automation Studio: hasnain-automation-studio.vercel.app
- toon-mcp-server: pypi.org/project/toon-mcp-server/