Likith Nadendla

Likith Nadendla

AI Engineer & Cofounder of SyncLabs — building production-grade agentic systems and RAG platforms at scale with a focus on retrieval pipelines, LLM evaluation, and regulatory-ready AI services.

Portland, OR
503-454-6925
job@likith.net

About Me

AI Engineer building production-grade agentic systems and RAG platforms at scale. Expertise in retrieval pipelines, LLM evaluation, and operationalizing AI services in regulated environments. Shipped Python services powering real-time agent reasoning, optimized prompt scaffolding for reduced hallucination, and integrated enterprise AI systems with measurable compliance wins. Hands-on RAG architecture (chunking, retrieval scoring, grounding), evaluation harnesses for regression detection, and prompt engineering for production. Strong cross-functional collaborator across backend, frontend, and AI teams, comfortable bridging applied AI and platform infrastructure with measurable throughput and latency improvements.

ML Systems Architecture

Architecting and deploying secure, cloud-native ML systems with automated pipelines and data workflows

Infrastructure & DevOps

Hardening infrastructure across AWS and Kubernetes with CI/CD guardrails and infrastructure-as-code

Compliance & Monitoring

Enforcing compliance for healthcare and retail with comprehensive observability and monitoring systems

Technical Skills

Cloud & DevOps

Programming

Databases & Data Ingestion

Frameworks & Libraries

Tools & Platforms

Infrastructure & Networking

Machine Learning & Modeling

Observability & Reliability

Other Skills

Projects

Projects

A timeline of selected projects delivering ML, analytics, and cloud-native systems with measurable impact.

SyncLabs Studio — AI Lip-Sync Video Platform

SyncLabs Studio (Co-Founder & Engineer)

Co-FounderVisit →

Co-founded and engineered an AI-powered lip-sync video platform built on MuseTalk v1.5, Whisper audio embeddings, VAE latent encoding, and BiSeNet face blending, running on GPU-accelerated cloud infrastructure.

Responsible for full-stack architecture, production deployment, and product development. Platform supports multi-language sync, pre-built AI avatars, and processes most videos in 2 to 3 minutes with zero local setup.

ClusterDuck IoT Mesh

Decentralized IoT Mesh Network for Disaster Resilience

August 2021 – December 2021Status: Completed

Implemented a decentralized IoT mesh communication network using the open-source Cluster Duck Protocol to simulate resilient connectivity during disaster scenarios.

  • Deployed ESP32-based IoT nodes running the Cluster Duck firmware to form a LoRa-based mesh network with Wi-Fi client access
  • Configured gateway nodes (Papa Duck architecture) to aggregate telemetry and forward messages to centralized endpoints via available backhaul connectivity
  • Evaluated mesh formation behavior including node discovery, routing, and automatic rerouting during simulated node failures
  • Tested network resilience under various failure scenarios to validate disaster communication capabilities

Daily Stack — Multi-Agent Productivity System

Daily Stack (Co-Creator, 2nd Place, Claude Code Hackathon)

Claude Code Hackathon2nd Place

Built multi-agent AI system with four specialized Claude agents (task creation, prioritization, scheduling, reminders) working in concert.

Implemented evaluation harnesses to measure agent coordination latency and user satisfaction. Won 2nd place at Claude Code Hackathon. Demonstrates agentic workflow design, prompt orchestration, and rapid iteration.

iOS Activity Monitor — Privacy-First Local Model App

iOS Activity Monitor (In Progress)

In Progress

A privacy-first iOS app inspired by macOS Activity Monitor that surfaces all active processes on iPhone, explains in plain language what each process is and why it’s running, and flags unusual activity—all powered by an on-device local model with zero cloud dependency.

Offline Text-to-Speech AI Model

Privacy-First Offline TTS System

Personal experimental projectStatus: In testing

Jetson Nano-based offline TTS pipeline optimized for privacy and sub-second inference.

  • End-to-end GPU-accelerated inference with TensorFlow/PyTorch on CUDA.
  • Custom-trained neural voices tuned for the target domain and edge constraints.
  • Efficient training/evaluation loop plus thermal/resource monitoring.
  • 100% offline operation—no cloud dependency for privacy-sensitive contexts.
  • Integrated into multi-cloud dashboards to track inference latency, accuracy, and device health.

Impact: proves production-quality voice synthesis on embedded hardware for secure deployments.

Jetson Nano TTS System Setup

Experience

Professional Journey

A timeline of my professional experience and career milestones in cloud technology and software development.

2024-Present

MLOps Engineer — Ayumetrix

Portland, OR
April 2024 — Present

Built and operated a HIPAA-compliant laboratory data and MLOps platform to ingest and normalize HL7 test results, enable reproducible ML-driven analytics, and automate secure reporting for partner labs and patients, with strict PHI isolation, auditability, and operational reliability.

Deployed production healthcare AI assistant (Nova) on AWS BedrockAgentCore using Claude 3.5 Sonnet with persistent multi-strategy memory; designed evaluation harnesses to catch hallucination before release
Integrated Claude 3.5 Sonnet streaming to auto-generate personalized health summaries and wellness recommendations per lab result; optimized prompt scaffolding to reduce hallucination and improve medical fact grounding
Built a vector knowledge base on Amazon Bedrock with automated FAQ ingestion; tuned chunking strategy and retrieval scoring to improve grounding heuristics
Architected an MCP gateway with Cognito M2M JWT auth to expose Lambda tools to Bedrock agent for secure real-time reasoning; designed evaluation harnesses to catch regressions on every release
Designed an event-driven PDF report pipeline with Claude integration, rendering branded AI summaries to S3 and catching hallucination failures before production
Managed multi-environment Terraform across eight repositories with layered dependency ordering; enforced compliance-aware audit logging for HIPAA workloads
Implemented OIDC-based GitLab CI/CD with test, validate, plan, approve, apply, tag, and notify workflows across all repos
Architected a zero-SaaS HIPAA observability platform spanning 14 AWS services with 59 Grafana alerts and two-tier escalation for operational reliability

2023

DevOps/MLOps Engineer — Core Defender AI

Portland, OR
April 2023 — February 2024

Designed and built a Python-based AI agent platform on AWS to power intelligent decisioning, personalization, and real-time recommendations at scale. The system focused on agent orchestration, system reliability, and production-grade deployments using containerized services, CI/CD pipelines, and strong observability.

Designed and shipped AI agent workflows orchestrating candidate retrieval, ranking, and response composition; built evaluation harnesses to measure end-to-end latency, throughput, and accuracy regressions
Built Python-based agent orchestration services coordinating embedding generation, retrieval, ranking, and response composition; implemented REST API interfaces for safe frontend/backend integration
Developed Python inference services on AWS ECS supporting high-throughput, low-latency agent reasoning; instrumented observability (metrics, logs, traces) to catch regressions and measure RAG quality
Implemented CI/CD pipelines to build, test, scan, and deploy containerized agent services across dev, staging, and prod with safe rollout strategies
Established observability for agent reasoning pipelines with evaluation harnesses catching hallucination and retrieval failures; defined SLIs/SLOs for agent latency, throughput, and accuracy
Collaborated with engineers and a principal architect to review designs, make architectural tradeoffs, and scale the system for higher maintainability

2019-2022

Linux System Administrator — Hexagon R&D India

Hyderabad, India
May 2019 — August 2022

Administered Linux infrastructure supporting GIS research environments and geospatial compute workloads

Administrated Linux servers for geospatial compute, optimizing system parameters for GIS data processing
Implemented user/group policies and RBAC for research teams, improving accountability and system stability
Migrated legacy desktop nodes to centralized Linux servers, improving maintainability and resource utilization
Provided tier-2 support for GIS modeling tools, build tooling, and CI pipelines; created troubleshooting documentation
Automated health checks, log analysis, and maintenance using Bash scripting, reducing manual intervention
Configured NFS mounts, repositories, and package mirrors; collaborated on OS hardening and security practices

Education

Academic Journey

My educational foundation in computer science, cybersecurity, and advanced technologies that shaped my career.

Master of Science in Computer Science

Pace University - Seidenberg School, New York City

Specialized in cutting-edge technologies and advanced computer science concepts with focus on practical applications.

Data Engineering & Analytics
Cloud Computing & Infrastructure
Machine Learning & AI
Software Development & Systems Design
Pace University Logo

Cyber Security Certification

Indian Dutch Cybersecurity School IDCSS

Specialized certification program focusing on international cybersecurity frameworks and strategic security analysis.

Cyber Diplomacy & International Relations
Threat Intelligence & Risk Assessment
Security Policy & Strategic Analysis
Cross-Border Security Coordination
The Hague Centre for Strategic Studies - IDCSS Partner
Government of Telangana - IDCSS Partner

Bachelor of Technology in Computer Science

Vardhaman College of Engineering (VCEH), Hyderabad

Comprehensive undergraduate program establishing strong foundation in computer science fundamentals and practical applications.

Software Development
Database Management
Security & Risk Management
Business Intelligence & Process Optimization
Vardhaman College of Engineering Logo

Blog

Coming Soon

Insightful articles about ML, DevOps, cloud architecture, and engineering best practices coming soon.

Get In Touch

I'm always interested in hearing about new opportunities and exciting projects. Whether you want to discuss cloud solutions, data analytics, or software development, feel free to reach out!

Location

United States