Open to opportunities

AhmedAli

LLM Inference Engineer · LLM inference optimization (vLLM, quantization, batching, KV-cache tuning) · Production ML systems on Kubernetes with GPU fleet management and autoscaling

Not specified, US
Book intro with Ahmed
LLM Inference EngineerML Platform EngineerML Systems EngineerModel Serving Engineer
01 — Positioning

Where Ahmed is headed.

Early-career ML systems engineer with nearly 3 years driving high-impact inference optimization: vLLM migrations, quantization strategies (INT8/AWQ), and autoscaling across GPU fleets. Balances hands-on infrastructure work (Kubernetes, AWS, Prometheus) with deep model-serving expertise while completing an M.S. in Computer Science full-time.

What Ahmed brings
  1. 01LLM inference optimization (vLLM, quantization, batching, KV-cache tuning)
  2. 02Production ML systems on Kubernetes with GPU fleet management and autoscaling
  3. 03Performance profiling and latency reduction (PyTorch Profiler, request tracing, observability)
  4. 04End-to-end ML deployment (model evaluation, load testing, on-call reliability)
  5. 05Full-stack capability (FastAPI, React, PostgreSQL—bridges ML and application layers)
02 — Proof

The receipts.

Numbers first. Every metric below is something Ahmed shipped or drove — not a claim, an outcome.

  • Proof 01
    30%
    GPU memory reduction, freed capacity for additional workload
    INT8 quantization deployment
  • Proof 02
    20%
    reduction in idle GPU-hours across ~40-GPU fleet
    Queue-depth autoscaling on EKS
  • Proof 03
    8%
    prediction accuracy improvement
    Root Stress Detection System (IoT + ML)
  • Proof 04
    10%
    end-to-end latency improvement via async worker pool
    Tokenization bottleneck elimination
  • Proof 05
    1.6×
    throughput increase, 25% p99 latency reduction
    vLLM migration and tuning
03 — Role fit

Operating range.

Where Ahmed plugs in fastest — sorted from strongest fit to adjacent. Not a wishlist; a routing map.

  • primary fit
    Inference/Model Serving Specialist
    Mid
  • primary fit
    ML Systems Engineer
    Mid
  • secondary fit
    Infrastructure/DevOps Engineer
    Mid
  • adjacent fit
    Full-Stack Engineer
    Mid