years in applied AI & creative tech
From early creative coding experiments to production-grade reasoning systems.
About
About Pete
Research engineer building dependable reasoning systems that balance creativity, reliability, and human delight.
Research & Systems
Pittawat Taveekitworachai (Pete)
I explore how large language models reason, create, and assist responsibly. My work spans research, product, and community—turning fragile ideas into dependable systems and sharing the playbooks that help others build with confidence.
Data points that anchor how I build, teach, and experiment.
published papers & deep dives
Exploring creativity, evaluation, and tools that stay reliable in production.
shipped projects & open experiments
Prototyping boldly, then hardening what works for partners and teams.
talks, workshops & keynotes delivered
Helping communities across APAC build responsible AI workflows.
Focus & Current Questions
Each project is a laboratory. Here are the questions that keep me iterating on research, infrastructure, and storytelling.
Creative intelligence without collapse
Testing how large models stay imaginative without drifting away from truth.
- When do hallucinations become a useful spark instead of a failure?
- How does model scale shift the boundary between imagination and reliability?
- Which safety rails keep generative systems playful yet grounded?
Orchestrating algorithms with LLMs
Pairing classical computing patterns with neural priors to tackle complex tasks.
- Which traditional algorithms resonate—or clash—with probabilistic reasoning?
- How do pipeline topologies change when an LLM becomes a black-box oracle?
- Where should symbolic structure and neural intuition interleave?
Evaluation without shortcuts
Designing resilient evaluation frameworks that keep up with rapidly evolving models.
- Can LLMs deliver structured output as tasks grow multi-dimensional?
- How do we observe degradation across scales, prompts, and modalities?
- What automation keeps humans focused on insight rather than grading?
Prompt architectures in practice
Turning orchestration heuristics into human-centered AI experiences.
- Why do prompt engineering patterns actually work?
- What does “exploration” look like when prompts become navigation in latent space?
- How can multimodal LLMs power creative tooling and healthcare support systems?
How I Build
Reliability, clarity, and accessibility thread through everything—from research pipelines to the tools communities adopt.
Applied research engineering
I build reproducible, resumable evaluation rigs that bridge notebooks, services, and production dashboards.
PythonTypeScriptPyTorchTransformersTensorFlowFastAPI
- Instrument experiments so findings stay auditable, extensible, and easy to reprise.
- Context-switch between numerical computing in MATLAB/R and modern LLM tooling.
- Use numpy, pandas, matplotlib, and seaborn to shorten the insight loop.
Full-stack systems & product
Ship interfaces and services that stay maintainable as teams grow and stakes rise.
ReactAstroNext.jsNode.jsDjangoGoJava
- Production experience across web, microservices, and API design, including LINE Shopping.
- Delivered mobile apps from native Android to React Native and Flutter rebuilds.
- Design background with Figma, Adobe XD, and Photoshop keeps user experience central.
Ops, assurance & platforms
Reliability stays non-negotiable—tests, automation, and observability are part of the canvas.
JestCypressPlaywrightSeleniumDockerGitFirebase
- Security and networking foundations inform threat models and review practices.
- Support rapid-response teams with structured workflows and clear runbooks.
- Work comfortably with SQL/NoSQL and vector stores—MongoDB, Neo4j, Pinecone, Milvus, Qdrant.
Trajectory
Trajectory Highlights
Research milestones, industry sprints, and community-led missions that inform how I frame the next question.
-
2025 — Typhoon Application Week
Built and shipped 7+ LLM-powered apps.
-
2025 — Typhoon T1 (research preview)
The first Southeast Asia reasoning model.
-
Oct 2024 – Present — Research Scientist, Typhoon team (SCB 10X R&D)
Large language model research and applications at SCB 10X.
-
Sep 2024 — M.Eng., Ritsumeikan University (Valedictorian)
Master of Engineering, Intelligent Computer Entertainment Laboratory.
-
2022 — Ethical Hacking Workshop
5‑day red‑team training by NIA & NBTC.
-
2022 — Research Associate (IC2, SIT KMUTT)
Led end‑to‑end ML/AR/web projects; presented Krathu‑500; hosted Kaggle competition.
-
2022 — B.Sc. (Computer Science), King Mongkut's University of Technology Thonburi (First Class Honour)
Bachelor of Science (Computer Science), Petch Prajom Klao scholarship; multiple academic awards.
-
2021 — LINE Thailand Internship
Full‑stack work on LINE Shopping; Vue 2→3 migration; microservices.
-
2020–2021 — Siriraj Hospital (Mahidol)
Tech lead & Flutter dev; React Native→Flutter migration; app size/load wins.
-
2020 — COVID‑19 Support Platform (KMUTT)
Led 10‑person team; 300+ shipments tracked; 50+ hospitals served.
-
2020 — Huawei ICT APAC (Network Track)
Participated in regional competition; routing/switching focus.
-
2019–2020 — Siriraj Hospital
React Native developer; added features and data collection flows.
-
2019 — Young Webmaster Camp 17th
Selected (programming major); built MVP for social impact.
-
2019 — Google Tech Challenge (Bangkok)
Selected team; coding + non‑coding puzzles.
Journey, Community & Service
Degrees and job titles matter, but the real story lives in the people, places, and service moments that shaped my judgment.
Education & Honours
M.Eng., Intelligent Computer Entertainment Lab, Ritsumeikan University
2022 — 2024Valedictorian focused on creative AI and evaluation tooling.
- Investigated generative creativity and reliability across large language models.
- Delivered Typhoon research contributions on reasoning and applied systems.
B.Sc. Computer Science, KMUTT (First-class honours)
2017 — 2021Petchra Pra Jom Klao Scholarship recipient and academic distinction awardee.
- Earned Jia Ra Nai Petchra Scholarship and Honor Award for program-leading performance.
- Name engraved on KMUTT’s Wall of Honors recognizing the top graduate of the cohort.
Research & Industry Collaborations
Applied research collaborations
Early in my journey I joined the Innovative Cognitive Computing Laboratory at KMUTT, partnering closely with clinicians and researchers.
- Led AR, web, and natural language projects that turned ideas into usable systems.
- Served as tech lead for Siriraj Hospital collaborations, shaping HFMD research tooling.
Industry immersion & mentoring
From LINE Thailand to national innovation programs, I thrive in teams that move fast without losing craft.
- Shipped LINE Shopping features and contributed to the Vue 2 → 3 migration across microservices.
- Selected for Young Webmaster Camp and Google Tech Challenge; mentored peers as inaugural Google Developer Student Club lead at KMUTT.
Service & Volunteer Leadership
- Directed an eight-person rapid-response team during the COVID-19 outbreak to build KMUTT’s nationwide donation management platform—300+ shipments, 50+ hospitals supported.
- Spearheaded Junior Programmer Camp XV, delivering admissions and staff tooling that opened computer science pathways for students across Thailand.
- Continue to review for conferences and coach communities because sharing expertise compounds impact.
Lifelong Learning Rituals
- Lifelong learning is ritual: I constantly take online courses, pairing curiosity with deliberate practice.
- Regularly tutor classmates and mentees—teaching is how I stress-test my own understanding.
- Connecting disparate topics keeps me inventive and empathic when designing intelligent systems.
Learning is how I stay curious and stay kind. Understanding a topic deeply makes it easier to design with empathy and to uplift the people building alongside me.