SohailGidwani
I build AI systems that actually work in production, not just in notebooks. Full-stack engineering meets applied machine learning.
Open to work · Remote · Hybrid · Relocation
0+
Years production experience
0
AI systems deployed at IIFL
0M
Parameter Custom Model
LA
Currently based
About
Building AI systems that work in production, not just in demos.
I'm an AI engineer and full-stack developer. I've worked on everything from RAG pipelines and LLM integrations to React frontends and Flask APIs. What I care about most is taking something complex and making it feel simple for the person using it.
I joined IIFL as a new grad and within months was building and deploying AI-powered apps used by real employees across the company: RAG chatbots, fraud detection, and automated support, all in production. Ambiguous problems don't scare me. If anything, those are the ones I find most interesting.
End-to-end ownership is how I work best. I built a multi-modal Alzheimer's prediction pipeline from data preprocessing through model training to calibration analysis, all as a single contributor. Shipping under pressure is where I've been tested, winning a hackathon at IIFL with an AI chatbot built in under 36 hours. Cross-stack fluency lets me move between PyTorch research code and production Next.js apps in the same sprint.
Education
Structured learning, layered over curiosity.
August 2025 - May 2027
Los Angeles, CA, USA
GPA - 3.75 / 4.0
M.S in Computer Science
University of Southern California
Advanced studies in AI systems, retrieval, and large-scale software engineering.
Coursework
Aug 2019 - May 2023
Mumbai, India
CGPA - 9.05 / 10
B.E in Computer Engineering
University of Mumbai - TSEC
Focus on AI/ML, cloud computing, and full-stack development with strong academic performance.
Coursework
Experience
Industry, research, and product execution.
Research Assistant
ActiveKeck School of Medicine of USC
Building MEMOIR-VLM, a multimodal deep learning pipeline for Alzheimer's disease classification and VQA using neuroimaging and clinical data, with end-to-end experimentation infrastructure across ~70M parameter models.
Senior Software Engineer - I
Insaito, Inc.
Joined while applying to MS programs. Received my USC admit the week I started. I chose to pursue the MS, communicated this openly with the team, and wrapped up cleanly.
Built an AI agent builder platform from scratch - open-source LLM deployment (Qwen 3, Mistral 24B), OAuth for 100+ apps, and MCP server development. End-to-end ownership from backend infra to frontend UI.
Full Stack - Software Developer
IIFL Finance Ltd
Built internal employee support chatbot, AI-powered fraud detection, and automated user support systems across multiple business verticals.
Skills
Organized by what you'd hire me to do.
- 01
Pythonlanguage - 02
HuggingFaceembeddings - 03
Qdrantretrieval - 04
Azurellm service - 05
Flaskapi - 06
SQLAlchemyorm - 07
Rediscaching - 08
Dockership
Full index
ML / AI / PyTorch · TensorFlow · Scikit-learn · Keras · OpenCV · NLTK · HuggingFace · Ollama · spaCy · Pandas · NumPy · Matplotlib · Seaborn · MCP · LLMOps
Languages / Python · JavaScript · TypeScript · Java · C++ · SQL
Web / React · Node.js · Express · Next.js · Django · Flask · Hono · FastAPI · Tailwind CSS · WebRTC · HTML5 · CSS3
Databases / PostgreSQL · MongoDB · Redis · Elasticsearch · MySQL · Qdrant · Oracle · Prisma · SQLAlchemy
Cloud / AWS · Google Cloud · Azure · Heroku · CloudFlare
Tools / Git · Claude Code · N8N · Docker · Kubernetes · CI/CD · Agile · RESTful APIs · Linux · Bitbucket · TFS · Serverless
Projects
Things I've built that I'm proud of.
Portage
An autonomous agent that migrates Flask apps to FastAPI and proves it with the repo's own tests.
LangGraph agent that plans a per-file task DAG, rewrites each file with an LLM, verifies in a network-off Docker sandbox, and recovers from failures under bounded budgets. Checkpointed to Postgres, so you can kill the worker mid-run and it resumes. Eval-proven on 6 pinned repos at K=3.
Wins
Certifications and wins that shaped my journey.
Beyond the Code
Stuff I care about when I'm not coding.
Marvel Universe
Spider-Man
No matter how beaten, how outmatched, he gets back up. Every time. That kind of resilience is something I try to carry into my own life.
Iron Man
Always has a backup plan. And a backup for the backup. If one thing goes south, there's already another plan ready. I try to think like that when I'm building systems.
Story-Driven Games
Simple Pleasures
Tap any to learn more
Let's workTogether.
Looking for full-time roles, research collaborations, or interesting side projects. Based in Los Angeles, open to remote and on-site.