Alex Sikand
Full-Stack AI Engineer
San Francisco, CA
AS
About
Full-stack engineer who ships production AI from architecture to revenue. Built CallSaver's voice AI platform end-to-end (100+ paying customers, sole engineer), shipped AWS infrastructure for 1,000+ enterprise dealerships at Impel, and productionized computer vision models at Silk Labs.
Work Experience

Jan 2025 - Present

Jan 2025 - Present
Founding Engineer
Remote
- Shipped voice AI platform from zero to 100+ paying customers as sole engineer — STT-LLM-TTS pipeline handling inbound calls, appointment booking, and after-hours routing for field service businesses.
- Built and deployed the full product surface: API, customer dashboard, marketing site, billing, job queues, and CRM integrations across separate staging and production environments.
- Owned architecture, deployment, and on-call from day one — debugging latency, tuning inference cost, and integrating webhooks with downstream CRMs.
TypeScriptNode.jsExpressZodOpenTelemetrySentryReact.jsViteAWSGCPRedis
BullMQTwilioIntercomLiveKitOpenAIClaude
LangfusePrismaPostgreSQL
BullMQTwilioIntercomLiveKitOpenAIClaude
LangfusePrismaPostgreSQL
Feb 2024 - Jan 2025

Feb 2024 - Jan 2025
Software Engineer
Remote
- Built and operated AWS infrastructure powering ServiceAI across 1,000+ car dealerships, processing 250K+ outbound customer messages daily via email and SMS.
- Maintained the omni-channel re-engagement platform driving dealership-customer reactivation and lifetime-value lift.
TypeScriptNode.jsAWSPostgreSQL
DynamoDBMongoDBTwilio

Jul 2022 - Jun 2023

Jul 2022 - Jun 2023
Software Engineer
Palo Alto, CA
- Built Python Django backend processing big-data telemetry from eVTOL aircraft for flight-data analysis.
- Modernized 30+ legacy UI components to current major versions and generated an OpenAPI spec covering 400+ API methods to unblock client integration.
PythonReact.jsTypeScriptMaterial UIOpenAPI

Sep 2020 - Jul 2022

Sep 2020 - Jul 2022
Machine Learning Engineer
Las Vegas, NV
- Led ML Ops at Silk Labs — built labeling pipelines for synthetic image data and training infrastructure on multi-GPU clusters.
- Productionized computer vision models for stadium-scale facial detection, firearm detection, and super-resolution — served 150+ concurrent RTSP camera feeds via Triton Inference Server with INT8 quantization.
PythonPyTorchComputer VisionTriton Inference ServerTensorRTWeights & BiasesDocker
Education

Boston University
M.S. Artificial IntelligenceMaster of ScienceArtificial Intelligence
2020 - 2021

Boston University
B.A. Computer ScienceBachelor of ArtsComputer Science
2016 - 2020
Relevant Coursework
CS 400Full Stack App DevelopmentCS 411Software EngineeringCS 440Artificial IntelligenceCS 542Machine LearningCS 565Algorithmic Data MiningCS 591Deep LearningCS 591Parallel ComputingCS 591User Centric Systems for Data ScienceCS 660Graduate DatabasesMA 225Multivariate Calculus
Skills
Languages
TypeScript
Python
Cloud & Infrastructure
AWS
GCP
Docker
Backend & APIs
Node.js
Express
Zod
OpenAPI
OpenTelemetry
Sentry
BullMQTwilio
Intercom
Frontend
React.js
Next.js
Vite
Tailwind CSS
shadcn/ui
Material UI
Databases & ORMs
PostgreSQL
MongoDB
Vector Databases
PineconeRedis
Prisma
Drizzle
AI / ML
OpenAI
Claude
MCP
PyTorch
LangChain
Triton Inference Server
TensorRT
Weights & Biases
LiveKit
LangfuseRAG
Computer Vision
My Projects
Check out my latest work
I've worked on a variety of projects, from simple websites to complex web applications. Here are a few of my favorites.

