ai engineering · full stack

I build AI agent workflows
and the evals that prove they work.

6+ years at Dell Technologies and TCS · M.S. Software Engineering, SJSU· via Accenture

Five public systems — agent evaluation, AI code review, multi-tenant SaaS, a Go time-series DB, and adversarial ML research.

Open to SWE & AI engineering roles · on STEM OPT

view projects get in touch resume

PythonTypeScriptJavaC#LLM evalsRAG

FastAPINext.jsSpring BootPostgreSQLRedisPyTorchGo

6 years

industry experience

Dell · TCS

10M+

Dell PCs

SupportAssist platform

2M+

active devices

services I owned · SupportAssist

eval fixtures

Sentinel · CI regression gate

featured work

Projects

01public

Archon

Autonomous agent in Python — planner/executor/reflector with a typed middleware chain (schema validation, retries, tracing on every tool call) and a statistical evaluation harness: seeded runs, per-step traces, and a failure-mode taxonomy.

PythonAgentsEval

case study repo

02public · MIT

Sentinel

GitHub App that reviews PRs via BM25 + dense (pgvector) retrieval, cost guardrails, and a 98-fixture eval harness — per-category P/R/F1, regression-gated in CI.

FastAPIpgvectorBM25

case study repo

03working study · MIT

Kairos

Multi-tenant OKR tracker that enforces tenant isolation in PostgreSQL with Row-Level Security, not in application code.

Spring Boot 3.3PostgreSQL 16Redis 7

case study repo

04public · single-node stable

Helios

WAL + LSM storage, Gorilla compression, label-postings filtering, a PromQL subset, streaming EWMA anomaly detection, and optional 3-node Raft replication.

GoLSM-treeGorilla

case study repo

05public · MIT

NeuroLens

MVP adversarial-ML research: from-scratch PyTorch models (no torchvision pretrained), FGSM and PGD attacks implemented from the original papers, defenses, and a multimodal-to-unimodal attack-transfer experiment.

PyTorchCLIP-lite

case study repo

Side builds worth a look

all repos

AdaptiveRateGuard

Distributed rate limiter in Go with a gradient-boosting predictor sidecar — proactively lifts limits before traffic spikes hit.

GoRedisgRPCGBM

experience

Commit history

2024 — 2026

San Jose State University

M.S. Software Engineering · Graduate Research

Graduate research in agentic AI systems: planner-executor agents with typed tool registries, evaluated on tool-call accuracy, schema adherence, and error recovery. Coursework in distributed systems, ML/NLP, and adversarial robustness. Built Archon, Sentinel, Kairos, Helios, and NeuroLens.

Aug 2021 — Jul 2024

Dell Technologies — SupportAssistvia Accenture

Senior Software Engineer

Embedded in the Dell SupportAssist product engineering team. C#/.NET microservices for Dell's preinstalled PC health platform: hardware diagnostics and scan orchestration, catalog-driven BIOS/driver/firmware update delivery with signature and hash verification, telemetry-driven predictive failure detection for batteries and storage drives, and the in-app AI Virtual Assistant experience with device-context injection and assistant-triggered actions. Re-architected long-running scan status from client polling to WebSocket push with resumable progress state.

Jun 2018 — Aug 2021

TCS

Software Engineer

Java Spring Boot microservices for an enterprise AI retail optimization platform: constraint-based space optimization across 1,400+ stores and 1,000+ parameters, distributed search with Redis caching sustaining sub-200ms P95 across 2M+ SKU records, multi-tenant OAuth2 SSO with tenant-scoped claims across product modules, and virtualized dashboard rendering for 100K+ SKU tables (~800ms to ~120ms).

contact

Open to SWE & AI engineering roles

M.S. Software Engineering, SJSU — graduated May 2026. On STEM OPT. Bay Area or remote. Prefer email — I reply within a day.

[email protected]resume PDF github

I build AI agent workflowsand the evals that prove they work.

Projects

Archon

Sentinel

Kairos

Helios

NeuroLens

Side builds worth a look

Commit history

Open to SWE & AI engineering roles

I build AI agent workflows
and the evals that prove they work.