Muhammad Umer Jamil

Full-Stack Backend Engineer & AI Automation Specialist

Based in: Lahore, PunjabCurrently: Backend Engineer, Technovate GlobalExperience: 2 years

Hire me

A Computer Science graduate from LUMS with extensive experience in building scalable REST APIs, multi-tenant SaaS platforms, and AI-driven automation workflows. Expert in FastAPI, Next.js, and Azure, with a proven track record of integrating complex third-party services like Stripe, Clerk, and OpenAI.

Experience

Backend Engineer

Technovate GlobalRemoteworkNov 2025 – Present

Summary

Designed and developed scalable REST APIs with FastAPI, supporting modular routing, JWT-secured endpoints, and production deployment on Azure App Service. Implemented Clerk-based authentication with JWT validation, user provisioning, and startup-scoped role-based access control (Owner, Admin) enforced via reusable permission dependencies and billing-aware account policies. Architected and optimized Microsoft SQL Server data models and query performance, encapsulating complex fundraising and CRM logic in T-SQL stored procedures. Built Gmail OAuth 2.0 integration for secure connect/disconnect flows, inbox read/send APIs, token refresh, and sent-email audit logging tied to investor pipeline workflows. Integrated Recombee for similar-deal recommendations, investor-startup matching, batch catalog sync, engagement tracking, and weekly recommendation persistence and email delivery. Leveraged Azure OpenAI and Azure Blob Storage to parse fundraising documents (PDF, PPTX, DOCX), extract structured deal fields, and generate investor-specific pitch-deck recommendations from guidelines and notes. Delivered investor relationship management features including Kanban fundraising pipelines, bulk import, team invitations with SMTP notifications, and multi-tenant startup context switching. Integrated Stripe Checkout, Customer Portal, and signed webhooks to manage subscriptions, promotions, and billing state synchronized with the application database. Developed a multi-tenant Data Room backend (FastAPI, SQL Server, Azure Blob) with secure document sharing, NDA workflows, password-protected guest access, and room configuration/audit APIs. Built and managed CI/CD pipelines with GitHub Actions (OIDC) for automated builds and deployments to Azure Web App across dev and production environments.

What I did

  • Built a fundraising SaaS platform using Azure OpenAI to extract structured deal information from PDF, PPTX, and DOCX files.
  • Implemented structured extraction, API integrations, batch syncing, engagement tracking, and recommendation persistence.
  • Integrated and evaluated LLM-based systems using Azure OpenAI for document extraction and recommendation workflows.
  • Worked on document extraction and AI-powered features.

Results

  • Automated manual document analysis and recommendation workflows to improve consistency and scalability.
FastAPIPythonSQL ServerT-SQLClerkRecombeeAzureApp ServiceBlob StorageOpenAIGoogle Gmail APIStripeGitHub ActionspytestJWT

Software Engineer

U.S.A based startupRemoteworkOct 2024 – Nov 2025

Summary

Design, build and maintain the software applications that are efficient and meet the business requirements. Map out complex business processes and translating them into efficient software and automated workflows. Develop robust no-code and low-code automation solutions. Focus on building business-friendly software, automating tasks, integrating systems, and optimizing workflows to save time and reduce errors. Work closely with stakeholders to identify development and automation opportunities and deliver solutions that add measurable value. Collaborated with various teams to deliver tailored software solutions, targeting full-stack development, API integrations, and user-centered design. Efficiently using, integrating, and decreasing the latency of 3rd party APIs to enhance performance and user experience. Built and deployed new features from the ground up, overseeing design, development, and delivery.

What I did

  • Mapped manual business processes and identified repetitive steps to build automated workflows.

Results

  • Turned a repetitive manual workflow into an automated, repeatable process by handling data transfer, API calls, notifications, and follow-up actions automatically.
  • Reduced human errors and improved process consistency by removing the need for manual information copying between systems.
NextJSReactJSExpressJSNodeJSSupabasen8nzapierplaywrightMongoDBNoSQLPostGreSQL

Skills

Technical

ReactJsReactJs
NextJsNextJs
Tailwind CSSTailwind CSS
FastAPIFastAPI
GitGit
VS Code
PythonPython
GithubGithub
JavaScriptJavaScript
HTML/CSSHTML/CSS
Zapier
SQLSQL
MS Azure
VercelVercel
TypeScriptTypeScript
NodeJsNodeJs
ExpressJsExpressJs
MongoDBMongoDB
OpenAI Integrations
n8n
FigmaFigma
C/C++C/C++
Visual Studio
Pandas
NumPy
NetlifyNetlify
Make.com
Framer
Matplotlib
AWS
Scikit-Learn
Canva
Seaborn
Tkinter
Azure OpenAI
Recombee
Image-based search
Real-time chat
Distributed Systems
Database Internals
Operating Systems
Software Engineering
ReactReact
Machine Learning Pipeline
Model Evaluation
Hyperparameter Tuning

Projects

MinarMarket

Full-stack Developerpersonal
Developed a full-stack marketplace platform by implementing the user interface on the frontend and the corresponding business logic on the backend to fulfill all use cases. Created MongoDB models to represent the key actors involved in the use case. Managed tasks using GitHub Boards, conducting sprint planning and retrospectives. Implemented real-time chat using sockets, AI-powered search, and image-based search for enhanced user experiences. Designed the UI in Figma with a focus on usability and visual appeal.
NextJsExpressJsNodeJsMongoDBJavascriptTailwind CSSSocketsFigma

Cephyron

Backend Developerpersonal
Built a full-stack SaaS platform with role-based access control (RBAC) supporting secure multi-user workflows (admin, client, staff). Integrated Google APIs for Gmail connectivity, enabling in-app email sync, read, and send functionality. Implemented AI-based document parsing and recommendation system. Integrated Stripe for subscriptions and billing, and designed a scalable backend using FastAPI and SQL with optimized database architecture.
FastAPISQL ServerClerkOpenAIRecombeeMS AzureStripeGoogle APIs

Matchstick

Co-developerpersonal
Co-developed of MatchStick's full-stack web application using NextJS and Supabase. Developed an AI feature in the app to parse the information from pdf, photos or videos to autofill the registration form by automatic mapping. Developed the voice recording and video meeting recording feature that transcribes the text to autofill the form through automatic mapping. Developed the feature for giving hashtags to the proposals, search them on basis of proposals and export the resultant profile cards as csv. Developed and upgraded already existing features on the frontend to allow better state management and filtering. Used Supabase and its built-in PostgreSQL for efficient data storage and retrieval.
NextJSSupabasePostGreSQLTypescriptOpenAINext.jsReact

Customer Churn Prediction Model

personal
Built a classification model to predict customer churn based on historical data during my degree. Handled the full pipeline including data preprocessing, feature selection, and model training.
Logistic RegressionRandom Forest

Academic Pretrained Model Experiments

personal
Experimented with pretrained models for smaller academic projects, including preparing datasets, training models, and evaluating their performance.
Pretrained models

Education

Bachelor of Science

Lahore University of Management Sciences (LUMS)Computer ScienceSep 2021 – May 2025

Recognition

Awards

BS Computer Science (CGPA 3.66)

LUMS

Graduated in 2025 with a 3.66 CGPA.

Hire MuhammadReach out about a role, a contract or a conversation.For recruiters

Muhammad's twin is AI, it can make mistakes.