96.06 · About · 10 records

Mariam Qadeem

AI Solutions | AI Agents & Automation | I build products that Think, Act and Scale

Everything in one place, no AI in the way. Jump to a section or just scroll.

Experience

Where I’ve worked

88.1 MHz

  1. No. 03

    2026

    Now showing

    Devsinc

    AI/ML Software Engineer

    Agentic scheduling platform for a healthcare client

    Next.js · NestJS · AWS Bedrock · LangChain · MCP · Retell · PostgreSQL · pg-boss

    Read more

    Core engineer on an agentic scheduling platform for a healthcare client.

    • Owned 5+ backend modules end to end in NestJS and Next.js on an agentic scheduling platform targeting ~70% automation.
    • Engineered a multi-stage slot-ranking engine for automated appointment offers.
    • Cut Retell voice-agent latency by ~40%.
    • Built 2+ production data pipelines and HIPAA-eligible LLM inference on AWS Bedrock.

  2. No. 02

    2025

    JUN 2025 — AUG 2025

    Clustox

    Software Engineer

    Features for EmailVerify, a SaaS email verification platform

    Ruby on Rails · REST APIs · ActiveRecord · CI/CD · Git

    Read more

    Software engineer on EmailVerify, a SaaS email verification platform.

    • Shipped 10+ features and bug fixes to EmailVerify, a production SaaS, in Ruby on Rails.
    • Built RESTful API endpoints and ActiveRecord models, shipped through CI/CD in Agile sprints.

  3. No. 01

    2025

    FEB 2025 — MAY 2026

    PUCIT (FCIT)

    Teaching Assistant

    Machine Learning, Application of ICT and Database courses

    Python · Machine Learning · Databases

    Read more

    Teaching Assistant for three courses: Machine Learning, Application of ICT and Database.

    • Machine Learning (Feb–May 2026): tutorials and labs on supervised and unsupervised learning, guided students implementing models in Python.
    • Application of ICT (Sep 2025–Jan 2026): prepared and delivered tutorials, labs and course material.
    • Database (Feb–Sep 2025): hands-on lab guidance, troubleshooting and assignment feedback.

Projects

Things I’ve built

01 / 06 · AI

Backstory

RAG memory layer for legacy systems

  1. Connect code, tickets and docs
  2. Ingest and embed
  3. Hybrid retrieval
  4. Evidence gate
  5. Cited answer
  6. Citation check or honest refusal
≥80%
eval gate: recall@5
≥90%
eval gate: citation correctness
View projectSource for Backstory (opens in a new tab)Live Backstory (opens in a new tab)
Illustration: a highlighted line of old code linked to a cited answer, what and why
  • Next.js
  • FastAPI
  • pgvector
  • Groq
  • Celery
  • Redis
  • PostgreSQL
  • Whisper
  • Clerk
  • pytest
  • Playwright
  • Docker

02 / 06 · AI

StreetLight

AI and blockchain civic complaint platform

  1. Citizen report
  2. AI classification
  3. Fraud check
  4. Community verification
  5. Auto-routing to department
  6. Citizen-confirmed resolution
92%
classification accuracy
5
AI pipeline layers
Score ≥ 85
auto-verify threshold
View projectSource for StreetLight (opens in a new tab)Live StreetLight (opens in a new tab)
Illustration: a street lamp over a pinned pothole, and a report, verify, resolve path
  • PyTorch
  • EfficientNet-B3
  • FastAPI
  • Flutter
  • React
  • PostgreSQL
  • Supabase
  • Solidity
  • Ethereum Sepolia
  • Firebase FCM
  • Hugging Face Spaces
  • Docker

03 / 06 · Data engineering

Lakehouse-Ledger

Governed medallion lakehouse on Databricks

  1. Public EIA API
  2. Bronze: raw and immutable
  3. Quality gate
  4. Silver: clean, with quarantine
  5. CDC and SCD2 history
  6. Gold KPIs and dashboard
View projectSource for Lakehouse-Ledger (opens in a new tab)
Illustration: bronze, silver and gold data layers with a quality check
  • Databricks
  • Delta Lake
  • Lakeflow Declarative Pipelines (DLT)
  • Unity Catalog
  • PySpark
  • Databricks SQL
  • GitHub Actions

04 / 06 · AI

SoundHive AI

Beehive health monitoring with machine learning

  1. Hive sensor data
  2. Time-series cleaning
  3. Window features
  4. Health labeling
  5. Model benchmark
  6. Neural network and demo
98.82%
neural network accuracy
0.88
F1-score
400,000+
sensor readings analysed
View projectSource for SoundHive AI (opens in a new tab)Live SoundHive AI (opens in a new tab)
Illustration: a honeycomb with one glowing cell above a sensor reading line
  • TensorFlow
  • XGBoost
  • scikit-learn
  • Streamlit
  • Python
  • Keras
  • Pandas
  • NumPy
  • Hugging Face Spaces

05 / 06 · Research

Claude Code Security

Research: AI vs. rule-based bug detection

  1. Ground truth fixed first
  2. Semgrep scan on vulpy
  3. Syntactic vs. semantic bugs
  4. Claude Code risks (CIA + CVEs)
  5. Hybrid pipeline design
  6. Coverage 39% → 52%
52%
coverage with the hybrid pipeline
6%
false-positive rate after the LLM filter
View projectSource for Claude Code Security (opens in a new tab)
Illustration: a code scan passing through two lenses, a rules grid and an AI lens
  • Semgrep
  • Claude Code
  • Python
  • Flask
  • OWASP

06 / 06 · Full stack

Aurora

Inclusive learning platform for diverse learners

  1. Student profile and needs
  2. Personalised lessons
  3. Quizzes with instant feedback
  4. Progress and study streaks
  5. AI chatbot support
View projectSource for Aurora (opens in a new tab)
Illustration: an aurora of soft light bands over an open book
  • Ruby on Rails
  • React
  • Tailwind CSS
  • PostgreSQL
  • Vite

About

A little about me

  1. Introducing06A

    Mariam Qadeem

    AI/ML Software Engineer · Lahore, Pakistan

    I'm an AI/ML software engineer who builds AI products from the first prototype to production: RAG systems, LLM integrations and agents, and the full-stack software around them.

  2. Education06B

    Bachelor of Information Technology

    PUCIT · 2022 — 2026

    CGPA 3.80/4.00, top academic rank, within the top 3% of students.

  3. Now06C

    2026

    Devsinc

    AI/ML Software Engineer

  4. Experience06D

    2025

    Clustox

    Software Engineer

  5. Experience06E

    2025

    PUCIT (FCIT)

    Teaching Assistant

  6. Skills06F

    AI/ML & Agentic Systems
    • TensorFlow
    • PyTorch
    • scikit-learn
    • NumPy
    • RAG
    • LLM Integration
    • LangChain
    • Agentic AI
    • Voice AI (Retell)
    Languages
    • Python
    • JavaScript
    • TypeScript
    • Java
    • C/C++
    • Ruby
    Frontend & Mobile
    • React
    • Next.js
    • Flutter
    • HTML/CSS
    • Tailwind CSS
    Backend & Databases
    • Node.js
    • Express.js
    • NestJS
    • FastAPI
    • Ruby on Rails
    • PostgreSQL
    • MongoDB
    • Firebase
    Data Engineering
    • Apache Kafka
    • ETL/ELT Pipelines
    • Databricks
    • Delta Lake
    • Lakeflow Declarative Pipelines (DLT)
    • PySpark
    • DuckDB
    • Airflow
    Cloud, APIs & Core CS
    • AWS (EC2, S3)
    • Docker
    • CI/CD
    • Git
    • RESTful APIs
    • JWT Auth
    • MVC
    • Design Patterns
    • DSA
    • OOP
    • Cross-Platform Dev

  7. Courses & recognition06G

    • Top 10, Codebees All-Female Inter-University Competition
    • Claude Code in Action (Skilljar)
    • Introduction to Agent Skills (Skilljar)
    • Claude 101
    • AI For Everyone (Andrew Ng)
    • Introduction to Software Engineering

  8. Say hello06H

    Thanks for watching.

    mariamqadeem006@gmail.com (opens an email in a new tab)

    GitHub ↗LinkedIn ↗

Contact

Give me a ring

mariamqadeem006@gmail.com (opens an email in a new tab)

GitHub (opens in a new tab)LinkedIn (opens in a new tab)