> tharaka.agent --status

I build AI agents that do real work, with humans still in charge.

now shipping:

Product leader building agentic and conversational AI for enterprises.

  • AI agents

  • Voice AI

  • Multilingual EN/SI/TA

  • Agent evals

  • Guardrails

  • RAG

  • ElevenLabs

  • Langfuse

  • AI agents

  • Voice AI

  • Multilingual EN/SI/TA

  • Agent evals

  • Guardrails

  • RAG

  • ElevenLabs

  • Langfuse

// case studies

Six AI products I’ve led.

01

Live platform

AI contact-centre platform

Telecom and financial-services clients need voice agents, agent assist, analytics and AI quality checks at 300 concurrent calls under 1.5s latency.

→ Owned features and the roadmap; wrote a telco technical proposal and project plan

→ Conceptualised a 9-Indic-language mixing model capable of handling complex CX flows for a major Indian bank.

→ Worked on agentic CX and AI voice agents for telephony systems.

Telco

Banking

02

Live

AI SDR agent for B2B sales

The old SDR flow was rigid, asked too many questions and didn’t convert.

→ PM and author of the FRD and screen specs; redesigned the workflow into Prospect, Research and Outreach

→ Shaped it into a modular multi-agent product that takes a lead from prospecting through research to personalised outreach.

B2B sales

03

In dev

Enterprise AI orchestration & governance platform

Scattered data and processes, and ungoverned AI use.

→ Helping shape the roadmap by aligning design, technical and business stakeholders.

→ Driving discovery and building the demos enterprises need, while managing product operations.

→ Integrating conversational AI into digital workers.

Enterprise operations

04

Live

Customer-support agent with LLM evals

FAQ answers on WhatsApp and a web widget, grounded in scoped knowledge, with hand-off to humans.

→ Owned requirements and the roadmap, including a test playground

→ Defined Langfuse eval dimensions: FAQ relevance, response quality, edge cases, out-of-KB questions and cost

→ Used LLM-as-judge to evaluate responses

Customer support

05

Pilot

Digital-human service kiosk

A tablet avatar for a Nordic municipality, with face, gaze and gesture detection and Norwegian voice and chat.

→ Product and project coordination

→ Built the Norwegian knowledge base, with separate visitor and staff knowledge bases

→ Wrote UI specs and architecture docs; internal testing reached 18 of a 30-person target

Public sector

06

Proof of concept

Crop-intelligence AI

Plantation groups need to predict crop health, disease, water stress and yield from satellite and drone imagery.

→ Shaped the solution

→ Shaped predictive disease management to catch outbreaks early and stop them spreading, alongside yield and financial forecasting.

→ Mapped 26 possible KPIs, working towards disease detection, water-stress monitoring, yield prediction and financial forecasting.

Agriculture

// about

I’m Tharaka, a product leader in Colombo, Sri Lanka, building agentic and conversational AI for enterprises: voice and chat agents, AI SDRs, and the governance that keeps them safe to use. I care about the unglamorous parts that decide whether AI works: clear ownership, real-world testing, and value you can actually measure.

AI agents

I design agents that act, not just chat: multi-step workflows with clear guardrails, cited reasoning and human review where it matters, measured with real evals rather than demos.

Voice AI

I build multilingual voice experiences (English, Sinhala, Tamil) that handle real customer tasks, tuned for latency, turn-taking and trust.

Product leadership

I take AI products from fuzzy idea to shipped scope, with sharp PRDs, phased plans and teams aligned on what matters first.

// contact

Building something with AI agents, or trying to get one past the demo stage? I’m always happy to compare notes.

© 2026 Tharaka Hettihamu · Colombo, Sri Lanka

Create a free website with Framer, the website builder loved by startups, designers and agencies.