MaheenAkhtar
Khan.
Building full-stack systems and AI-powered products used by 5,000+ users across 500+ US auto dealerships.
The story behind
My journey.
Hey, I’m Maheen.
I’m a software engineer currently at PAM (remote), where I build full-stack systems and AI-powered call handling pipelines serving over 5,000 users across 500+ US auto dealerships.
My work spans the whole stack — backend APIs, LLM integrations, and the interfaces people use every day. I care about reliability and latency as much as I care about the product feeling right. Beyond the code, I stay close to what customers actually need: owning prompt engineering, evals, and cross-platform triage to keep things running smoothly.
I hold a Software Engineering degree from NUST (graduated May 2026). I’m also a UAE Golden Visa holder — based in Abu Dhabi, open to relocation.
1.9+
yrs
exp.
Where I’ve worked
Experience.
- –Shipped key features across the admin dashboard, call routing, and Agent Profiles systems — serving 5,000+ users and 500+ US auto dealerships
- –Built a configurable multi-step call routing system now deployed across all 500+ dealerships — cutting routing-related support requests by ~30%, reducing setup time from days to under an hour, and directly addressing a top onboarding request by enabling full receptionist-mode configuration
- –Integrated LLM inference pipelines (GPT 4.1, OSS-20B, OSS-120B) via prompt design and API orchestration, improving service matching accuracy by 24%
- –Engineered self-serve onboarding tooling that replaced manual DynamoDB edits with a guided dashboard flow — cutting onboarding time by 50% and removing engineering dependency from ops workflows
- –Own customer support end-to-end: prompt engineering, evals & benchmarks, cross-platform triage (prompts, integrations, dashboard, core platform), and developing custom solutions — helping retain high-value accounts including Parkway Family Auto Group
- –Built proactive Slack alerting for critical database changes and client feedback signals, cutting average issue response time by ~15% and eliminating the need for manual log monitoring across the team
- –Collaborated across a cross-functional async team — backend: Node.js / Express / Serverless / AWS (Lambda, DynamoDB, S3, SQS, SNS, SES, CloudWatch) / PostgreSQL / Telnyx / Twilio / Retell AI; frontend: React / Next.js / TypeScript / Mantine / Chakra UI / Radix UI / Tailwind; LLMs: OpenAI / Anthropic / Groq / Gemini
Things I’ve built
Projects.
Click any project to explore details
Academic background
Education.
Bachelor of Software Engineering (BESE)
National University of Sciences & Technology
CGPA
3.32 / 4.0
O / A Levels
The International School of Choueifat
Recognition & wins
Achievements.
Jun 2026
Shortlisted for Rector's Gold Medal Presentation
NUST SEECS · 40+ competing teams
May 2026
3rd Best Industry Project of Software Engineering
NUST SEECS Open House 2026 · 40+ competing teams
Jan 2024
UAE Golden Visa Holder
Government of the United Arab Emirates
AI, Cyber Hackathon & Drone Swarm Gala
20+ participating teams
NUST GPA-Based Scholarship Award
1-time recipient · National University of Sciences & Technology
Finalist — PAM Track
Headstarter Hiring Hackathon 2 · 30+ competing teams
Finalist — OloStep Track
Headstarter Hiring Hackathon 1 · 50+ competing teams
Top 20 Teams — 48-hour Hackathon
Headstarter AI · 550+ competing teams
Courses & credentials
Certifications.
Machine Learning Specialization
Complete JavaScript Course
Complete Web Development Course
Let’s connect
Get in touch.
If you’re working on something interesting, or just want to chat — I’d love to hear from you.