AI Systems Engineer
Bridging the gap between high-level Generative AI and bare-metal compute optimization. Building systems that are blisteringly fast, scalable, and highly cost-efficient.
Agents: Planning, Orchestration & Testing
Often operating as an AI consultant, I deeply analyze regulated, complex business workflows to architect and deploy robust agentic automation strategies.
Moving beyond simple chatbots, I build sophisticated multi-agent systems with mandatory Human-In-The-Loop (HITL) checkpoints and execution traceability to ensure enterprise-grade trust.
To guarantee reliability, I construct rigorous evaluation pipelines, generating synthetic datasets from historical organizational data to systematically optimize prompts and routing logic.
Complex Orchestration
Designing stateful multi-agent workflows to automate highly regulated, multi-step organizational processes.
Human-In-The-Loop
Implementing explicit authorization gates and transparent reasoning traces to build trust and ensure safety.
Rigorous Evaluation
Synthesizing domain-specific test datasets to empirically benchmark and optimize agent performance before production.
GPU Optimization & Scalable AI
Moving beyond basic API wrappers to engineer production-ready AI infrastructure. At Intuitive, I architected an enterprise RAG framework that served 15+ applications while strictly managing compute overhead.
By routing workloads to on-prem GPUs and tuning concurrent processing, I drastically reduced reliance on expensive cloud LLM APIs. The core focus was pushing the hardware limits: optimizing GPU memory allocation, batching requests, and leveraging multi-threading to achieve massive throughput gains.
75% Latency Reduction
Slashed chat latency from 90s to 20s via multi-threading and optimized batched inference.
4x RAG Throughput
Tuned vector database retrieval and generation pipelines for high-concurrency environments.
On-Prem Deployment
Deployed adaptive LLM inference services on Kubernetes to efficiently manage local hardware resources.
Systems Level Engineering
Memory Management
Fine-tuned KV caching and continuous batching algorithms to maximize GPU VRAM utilization without triggering Out-Of-Memory (OOM) errors under heavy load.
Distributed Workloads
Built load-balancing layers across Kubernetes clusters to distribute inference tasks dynamically based on real-time node availability and token processing speed.
Cost Efficiency
Replaced dependency on per-token cloud pricing with fixed-cost on-prem hardware architectures, heavily optimizing the cost per query for enterprise-scale deployments.
Bare-Metal Edge Optimization
Executing AI models on edge devices like Jetson Nano and UAV hardware requires moving beyond Python scripts into deep systems-level engineering.
At Dronaid, I ripped out inefficient Python inference pipelines and rewrote the core perception modules in C++. By compiling models down to low-level execution engines, I eliminated overhead and doubled hardware performance for mission-critical flight tasks.
Inference Acceleration
More than doubled YOLO real-time inference from 5 FPS to 12 FPS on edge UAV hardware.
Low-Level Tooling
Utilized TensorRT, NVIDIA Triton, and CUDA to optimize neural network layer execution.
Concurrency & MPI
Engineered multi-threaded C++ architectures (OpenMP) separating capture, preprocessing, and inference.
Building from Zero to One
I spend my weekends building power-user apps to automate and optimize my daily life. From self-hosting wealth trackers to deploying personal AI agents, I love taking a product from zero to one.
