Systems Engineer – Inference & Database Internals, based in Guwahati, India
Currently working at SmatBot
class SystemsEngineer:
def __init__(self):
self.languages = ["C++", "Python", "Rust", "Metal (MSL)"]
self.systems = ["Buffer Pools", "B+ Trees", "Concurrency Control", "GPU Kernels"]
self.infrastructure = ["FastAPI", "Docker", "AWS", "Linux"]
self.current_work = {
"role": "AI & Systems Engineer @ SmatBot",
"focus": "Inference latency optimization & high-throughput data pipelines"
}
self.open_source = {
"llama.cpp": "Metal GPU backend — GGML_OP_CONV_3D (F16/F32) & Gated Linear Attention kernels",
"Docling (IBM)": "Re-engineered table detection with Flood Fill (BFS)",
"LanceDB": "Sync/Async API parity in Python SDK"
}
def deep_tech_projects(self):
return [
"CMU BusTub — Buffer Pool Manager, B+ Tree Index, Query Execution (C++)",
"Metal kernel development for inference operators (MSL/GGML)",
"Production inference pipeline re-architecture (25-35% latency reduction)"
]
def currently_exploring(self):
return "GPU compute, inference optimization, and database internals."



