Lasantha Kulasooriya

Machine Learning Engineer Open to remote AI / ML roles

I build AI agents that decide and act, not just answer.

I design and deploy production AI agents across orchestration, retrieval, and real-time voice, alongside computer vision at the edge. These systems run in front of real users in cloud and physical environments, so latency, recovery, and failure modes are design inputs, not afterthoughts.

Download résumé
Scroll

Selected Work

05 / projects
01
Behavior Orchestration Agent
C++ ROS 2 BehaviorTree.CPP agent that decides what the Manriix robot does next — navigate, pause for a person, dock to charge, or ask for help — and recovers cleanly when any of those fail. 20 documented architecture decisions and a self-authored production-readiness audit closed 5 blockers before it shipped.
C++17ROS 2BehaviorTreeAgents
02
Cloud-Edge Voice AI Platform
Voice agent that runs on the robot's own Jetson hardware but is configured, deployed, and monitored from the cloud — LiveKit, Deepgram, OpenAI, and ElevenLabs behind one agent runtime, with MQTT deployment sync so a live session never blocks on a cloud read.
PythonLiveKitVoice AIMQTT
03
gst-nvdsvpistereo
Open-source GStreamer DeepStream plugin for VPI stereo depth on NVIDIA Jetson Orin — sustained 26 fps (up from 21) via the OFA+PVA+VIC hardware pipeline. Validated against tape-measure ground truth from 2 to 12 meters before it ran in production.
C++DeepStreamVPIJetson
04
E2E Financial Data Extraction
FastAPI + React pipeline turning messy financial PDFs into validated structured data, entirely offline — LayoutLMv3, Table Transformer, and a locally-run Mistral-7B, 85–95% line-item accuracy.
PythonLLMFastAPIRAG
05
Multi-Camera Navigation Behavior
Direct ZED SDK perception across three cameras, BIRCH clustering, and a multi-factor scoring function decide where the Manriix robot should stand near a group of people — ~19 fps sustained across all three cameras at 20–30% GPU utilization.
PythonROS 2ZED SDKCUDA/CuPy

Who Am I

about
Lasantha Kulasooriya
Based in
Colombo, Sri Lanka
Currently
Senior ML Engineer, Hype Insight Global
Focus
AI Agents · Applied LLM Systems · Edge AI
Availability
Open to relocation · remote, hybrid, or onsite
Education
BSc (Hons) Data Science, Univ. of Peradeniya

..Most AI agent engineers have never shipped a system where a wrong decision has physical consequences.
I have.

Since 2024 I've worked at Hype Insight Global, an Australian robotics company with a subsidiary office in Colombo, on Manriix, an interactive service robot. I wrote the BehaviorTree.CPP mission logic that decides what the robot does next, built its multi-camera perception pipeline, and, most recently, a cloud-edge voice AI platform that lets it hold a real-time conversation.

That work set the bar for what "reliable" means to me. A behavior tree can't retry indefinitely, so I bring that same discipline to AI agents: systems that call tools, retrieve context, and act on it correctly, not just generate plausible text.

I started in insurance data science at Janashakthi Insurance, building a claim-prediction model that outperformed the baseline before I ever touched a robot. Today I maintain gst-nvdsvpistereo, an open-source Jetson stereo-depth plugin I validated against physical tape-measure ground truth before trusting it in production. That's the discipline I've built across both halves of my career: document decisions as I make them, verify before I trust a result, and treat reliability as a design input, not an afterthought. It's the same standard whether it's a robot operating next to a person or an agent deciding which tool to call. Outside of that, I'm usually at the gym, making chicken curry, or watching a show.

Where I go deep

Agents & Orchestration

Autonomous decision logic and multi-agent systems: behavior trees, LLM agents, production reliability.

Physical AI & Robotics

ROS 2 systems that run on real hardware: safety behavior, navigation recovery, human-aware robot control.

Applied LLM Systems

RAG pipelines and document intelligence wired into real workflows, not notebooks.

Voice & Realtime AI

Low-latency STT/TTS, voice interruption handling, and cloud-edge session architecture.

Edge AI & Perception

Real-time computer vision on Jetson: stereo depth, multi-camera fusion, TensorRT and CUDA optimization.

AI Platform Engineering

Cloud-edge architecture, production backends, and deployment sync that never blocks a live session.

Stack
AI Agents & Orchestration
BehaviorTree.CPPROS 2LangGraphLangChainMulti-Agent SystemsTool OrchestrationFinite-State Machines
LLM & Generative AI
OpenAIAnthropicMistralOllamaLiteLLMHugging Face TransformersRAGLangfusePrompt Engineering
Voice & Realtime AI
DeepgramElevenLabsLiveKitSpeech-to-Text / Text-to-Speech
Edge AI & GPU Computing
NVIDIA DeepStreamTensorRTCUDAJetsonGStreamerONNX
Computer Vision & Perception
OpenCVZED SDKRealSenseYOLOMediaPipeStereo Depth3D Vision
Backend & Data
FastAPIPostgreSQLMongoDBMQTTWebSockets
Languages & DevOps
PythonC++TypeScriptDockerGitCI/CD

Journey

experience & education
2026 — Now
Senior Machine Learning EngineerWork
Hype Insight Global · Robotics & Physical AI

Own the AI and robotics architecture behind Manriix, from behavior-tree decision logic to the voice platform it runs on.

  • Wrote the BehaviorTree.CPP mission executor coordinating the robot's safety, voice interruption, navigation recovery, docking, and operator-assist behavior, with atomic mission-state persistence so an interrupted session resumes cleanly on restart.
  • Designed the cloud-edge architecture for the robot's voice AI platform, LiveKit, Deepgram, OpenAI, and ElevenLabs behind one agent runtime, with MQTT deployment sync so a live conversation never blocks on a cloud read.
  • Built a Jetson AGX Orin voice-guided photo capture system combining YOLO11-pose, MediaPipe, and RealSense depth to position people and trigger Canon camera capture through gimbal control.
  • Documented 20 architecture decisions and ran a self-authored production-readiness audit before shipping, closing 5 blocking issues without being asked to.
2024 — 2026
Data ScientistWork
Hype Insight Global · Edge AI & Perception

Built the perception and camera systems that let Manriix operate autonomously around people.

  • Built the multi-camera perception and navigation-decision pipeline, sustaining ~19fps detection across 3 ZED X cameras at 20–30% GPU utilization, with BIRCH clustering and an 8-level recovery architecture for degraded perception or navigation states.
  • Built the robot's camera-control subsystem, integrating Canon R6 Mark II capture, video recording, and live streaming with hardware-accelerated encoding to keep camera I/O reliable under concurrent load.
  • Built the photo-retrieval service that watches capture sessions in real time and hands processed images off to the robot's frontend.
2024
Data Science & AI EngineerWork
Janashakthi Insurance (JXG) · Intern → Associate

Applied predictive modeling and analytics to insurance operations, from claim risk to real-time KPI reporting.

  • Built a deep learning claim-prediction model that outperformed the baseline.
  • Designed a personalized policy recommendation system for insurance agents; became my undergraduate thesis.
  • Built the real-time KPI pipeline and leaderboard for DriveMe 2024, the company's annual sales competition.
2019 — 2024
BSc (Hons) in Data ScienceEducation
University of Peradeniya · GPA 3.65
👋 Hi, I'm a digital version of Lasantha. Ask me about my work.