15+
Years Shipping Software
Software Engineer · AI Inference Enthusiast
Abdelrahman Selim
I design and ship production software with a focus on AI inference that is cost-efficient, fast, and reliable at scale.
Based in Riyadh with Saudi Premium Residency. I collaborate with teams across KSA and MENA to move from idea to dependable production systems.
Production AI Inference
Primary Focus
KSA · MENA · Remote
Region
Flagship Work
Impact-focused engineering projects
Inference Cost and Throughput Optimization
- Problem
- High serving cost and unstable latency under burst traffic.
- Constraints
- GPU budgets, production SLAs, and mixed model workloads.
- Role
- Led optimization strategy and rollout for inference serving.
- Solution
- Applied quantization, dynamic batching, and KV-cache tuning with production-safe guardrails.
- Impact
- Lower serving cost and improved response consistency under load.
Edge-Ready AI Deployment Pipeline
- Problem
- Model promotion to edge environments was slow and operationally risky.
- Constraints
- Multiple deployment targets and strict reliability requirements.
- Role
- Designed and implemented deployment automation strategy.
- Solution
- Introduced environment-aware deployment workflow with validation checkpoints and rollback paths.
- Impact
- Faster, more predictable releases with reduced deployment risk.
Scalable Serving Reliability Improvements
- Problem
- Service quality degraded during concurrent requests at peak periods.
- Constraints
- Need to preserve reliability while scaling request handling.
- Role
- Owned reliability hardening for critical serving paths.
- Solution
- Refined concurrency controls, tuned request routing, and improved failure-handling paths.
- Impact
- More stable service behavior under high concurrency conditions.
Trajectory
Career timeline
-
Now
Production AI Inference Focus
Building cost-efficient, fast, and reliable inference systems for real-world workloads.
-
Recent Years
Performance and Reliability Engineering
Delivered architecture and execution improvements across serving, deployment, and operational stability.
-
Career Foundation
Full-Stack to Systems Depth
Built a strong engineering base by shipping production software across multiple stacks and domains.
Capabilities
Skills and tooling focus
AI Inference
- Quantization
- KV-Cache Strategies
- Batching
- Model Serving
Systems
- Performance Tuning
- Reliability Engineering
- Edge Deployment
- Observability
Delivery
- Architecture Design
- CI/CD
- Release Safety
- Cross-Functional Execution
Next Step
Let us build production-grade software that delivers measurable outcomes.
If your team is working on AI systems, platform reliability, or high-performance delivery, I would be glad to discuss collaboration.