Startups shipping their first AI feature, scale-ups hardening one for production, and enterprise teams that need senior hands on an agentic/RAG/voice build. If "make this reliable, fast, and cost-sane" is the problem, that's the work.
AI assistants now recommend people and services. I help you become the one they cite — the same setup that brought a client to this very site. GEO / AI-search optimization →
What I Help With
Hands-on engineering, plus the architecture judgment behind it
Agentic AI & LangGraph
Design and build production LLM agents — orchestration, tools/MCP, memory, guardrails, verifiers, evals, and observability that hold up with real users.
RAG pipelines
Retrieval that actually grounds answers: chunking, embeddings, pgvector, hybrid search, semantic caching, and retrieval evals you can measure.
Voice AI
Realtime voice agents on OpenAI Realtime + WebRTC — sub-1.5s latency, barge-in, and cost engineering, drawn from shipping a live voice tutor.
AI architecture & design review
Bring your in-progress system; get a focused audit of orchestration, retrieval, evals, cost, and reliability with a prioritised action list.
GEO / AI-search optimization
Get your site found and cited by ChatGPT, Claude, and Perplexity: llms.txt, entity-graph schema, FAQ markup, AI-crawler access, and IndexNow. I did it for this site — and landed a client from it.
Fractional CTO
Part-time engineering leadership for AI-first products: strategy, architecture, hiring, and keeping the build on track.
Embedded & due diligence
Work alongside your team a few days a week, or assess an AI product/team for investors and acquirers.
Explore Consulting Services
Dedicated pages for each engagement — dive into the one you need
Voice AI Development →
Realtime voice agents on OpenAI Realtime + WebRTC — sub-1.5s latency, barge-in, STT/TTS, cost engineering.
MCP Development →
Custom Model Context Protocol servers and agent-to-systems integrations, with production-grade permissions and evals.
GEO / AI-Search Optimization →
Get found and cited by ChatGPT, Claude, and Perplexity: llms.txt, entity-graph schema, FAQ markup, IndexNow.
AI Architecture Mentoring →
1:1 mentoring and design reviews for senior engineers and teams building real AI systems.
Hire an Agentic AI Developer →
Freelance / contract agent builds — end-to-end, embedded, or rescue of a stuck project.
For Business Leaders
Outcome-focused engagements in plain language — no jargon required
Fractional CTO for Startups →
Part-time technical leadership — AI product strategy, architecture, and hiring — without a full-time CTO cost.
Reduce Costs with AI →
Cut operational costs by automating repetitive work — assessed and measured honestly.
AI Adoption for Business →
From AI readiness to your first high-value project — strategy for non-technical leaders.
Engagement Models
Shaped to the problem and your stage
Project
Fixed scope, clear deliverable.
- Ship a defined agent/RAG/voice build
- Milestones and a working system
- Handover + docs
Embedded / Forward-Deployed
Ongoing senior capacity — a forward-deployed engineer inside your team.
- Set days per week with your team
- Architecture + hands-on code
- Flexible month to month
Advisory / Fractional CTO
Leadership without a full-time hire.
- Strategy, architecture, hiring
- Design reviews and unblocking
- Investor/board-ready clarity
Where I've Shipped
15+ years of production systems — the track record behind the advice
Amazon
SDE II, Payments — hardened payment security, cutting CSRF impact 19.7%; built a DynamoDB throttling system that cut coupon misuse 15.3%.
Agoda
Senior Software Developer, Bangkok — Scala + Cassandra + Kafka A/B testing pipeline, 22% faster product insights.
CoinSwitch Kuber
Lead Developer — migrated the monolith to Python/FastAPI microservices; held the system stable from 600K to 10M users.
Wayfair (via ForaySoft)
Senior Engineering Manager — led 10 engineers on a seller-promotions migration; cut system failures 28%, supported +25% peak volume.
PhysicsWallah
Director of Engineering — hired and led 15 engineers building an in-house LMS; CI/CD lifted deployment frequency 15%.
Porch.com
Contract SDE, Seattle — built the Facebook Marketplace integration; doubled service-quote requests and lifted revenue 13%.
Earlier
Yatra, Sapient, CPA Global, Domino and Hadoolytics Technologies — enterprise and product engineering across Java, Spring, AWS, and distributed systems.
15+ years shipping production systems — Amazon (Payments), Agoda, and CoinSwitch (scaled 600K → 10M users) — and now building Yuvan, a live voice-first AI tutor on LangGraph, RAG, and OpenAI Realtime. You get senior engineering judgment grounded in systems that actually run.
Frequently Asked Questions
What kind of AI consulting do you offer?
Hands-on AI engineering — building agentic systems (LangGraph), RAG, and voice AI — plus architecture and design reviews, Fractional CTO, and embedded support. The focus is shipping reliable systems, not slide decks.
Do you build, or only advise?
Both. Most engagements are hands-on, but advisory-only (reviews, Fractional CTO, due diligence) is available when that's what you need.
What engagement models are available?
Project-based, retainer/ongoing advisory, embedded a few days a week, or Fractional CTO — whatever fits the problem and your stage.
Are you available remotely and worldwide?
Yes — remote-first across US, European, and Asian time zones, with short-term on-site possible for kickoffs.
How do we start?
Book a free 30-minute intro call to scope it; you'll get a clear recommendation and proposal.
Related
Training →
Upskill your team on agentic AI, LangGraph, RAG, MCP, and voice — 1:1 or corporate.
Technical Interview Services →
Hire an interviewer for your panel, or practise with mock interviews.
Speaking →
Talks, workshops, and guest appearances on agentic AI and production ML.
Yuvan & Projects →
See the production systems behind the advice.
Experience & Resume →
15+ years across Amazon, Agoda, CoinSwitch, and startups.