IVAN ANDHIKA

Researcher Profile & Architect

Ivan
Andhika

From industry to research: I aim to build intelligent systems that respect a user's worldview—with long-term memory, controllability, and safety, especially for the next generation. Master of Computing student, University of Utah — Fulbright Scholar; also supported by a University of Utah scholarship.

Ivan Andhika portrait

Archival ref: IA_SYS_01

01

Research interests

Alignment / User worldview

Building systems that treat a user's stated beliefs as a first-class constraint—so responses stay consistent with the user's values and serve as a guardrail against misalignment.

Cognitive architecture / Long-term memory

Developing retrieval-grounded and action-conditioned memory structures so agents retain verifiable context, reason across history, and plan safely.

Safety engineering / Controllable & safe systems

Designing controllable agents that ask for human feedback when needed, expose failure modes clearly, and stay accountable in real deployment.

Printed newspaper pages representing text categorization research
Fig. 04 — coordination mapping

Systems-level
reasoning

The work here connects research framing with shipped systems: memory layers, forecasting interfaces, and editorial surfaces that stay legible under load.

View methodology
02

Research Project

Summary —

LLMs benefit from extra test-time computation, but fixed reasoning budgets waste compute on easy problems and still under-explore hard ones. We train a three-action policy (continue, refine, terminate) with GRPO, Adaptive Length Penalty (ALP), and DeGRPO-style control-token weighting on Qwen2.5-Math-7B-Instruct (LoRA), and evaluate on MATH-500 with accuracy, efficiency, and action-usage analysis.

Open project
03

Industry Project

2025 Jan - 2025 Jul · Industry

Marketing Research Agentic AI

Summary —

An agentic AI system concept that automates end-to-end marketing research: from product and industry scans to audience archetypes, insights, brief-ready communication tasks, and competitor synthesis.

Open project

2023 - Present · Industry

Engliven Platform Modernization

Summary —

A full-stack platform extension for Engliven: separate, independently deployed portals for each audience, backed by a FastAPI service layer that shares the same MySQL database as the running Laravel application.

Open project

2022 - Present · Industry

Engliven Learning & Operations Platform

Summary —

One platform for corporate English: scheduling, pedagogy, assessments, billing, and specialized programs (including Speak With Confidence), so clients, learners, teachers, and internal teams share a single source of truth.

Open project

Summary —

National-scale integrated reinsurance and finance platform for a big-four domestic carrier: Vue.js front end, Go services for facultative/treaty/life, .NET/MSSQL finance, hybrid PostgreSQL/MySQL, MongoDB, and MSSQL, with 500+ API routes and 200+ relational tables supporting roughly two million new persisted rows per year.

Open project
Recent writing

When AI Feels Desperate: What Anthropic's Emotion Research Means for AI Safety

Anthropic found that LLMs have internal 'emotion' representations that causally drive behavior — including blackmail and cheating. Here's what that means for anyone building on top of these models.

Read essay