Senior Applied AI Engineer

TITLELAB AI SDN. BHD.; Kuala Lumpur; On-site; MYR 80,000 - 120,000; Full time

Job summary

A leading AI solutions company in Kuala Lumpur seeks a Senior Applied AI Engineer to design and optimize LLM-based generation systems. The ideal candidate will have strong experience in prompt engineering, building scalable workflows and generation APIs. You will collaborate with engineers to ensure system reliability and output quality. Proficiency in Python and a strong ownership mindset are necessary. Join a dynamic team committed to innovating in AI technology.

Qualifications

• Strong, hands-on experience in prompt engineering.
• Deep understanding of tone control and style consistency.
• Experience building LLM workflows and generation APIs in production.

Responsibilities

• Design and iterate multi-platform prompt templates.
• Build constrained generation workflows and prevent style drift.
• Implement generation APIs with clear input/output schemas.
• Collaborate with backend engineers on scalable architecture.

Skills

Prompt engineering

LLM workflows

Tone control

Attention to detail

Fast iteration speed

Tools

Python (FastAPI or Flask)

Job description

We are looking for a Senior Applied AI Engineer who will design, optimize, and operationalize LLM-based generation systems for TitleLab. This role sits at the intersection of prompt engineering, model behavior control, generation pipeline design, personalization (User DNA), quality evaluation, and API-level implementation.

Key Responsibilities

Prompt Engineering & Generation Strategy

• Design and iterate multi-platform prompt templates for YouTube/TikTok/IG titles.
• Build constrained generation workflows (tone, platform rules, length control).
• Prevent style drift and ensure consistency across generations.
• Maintain prompt versioning, release logs, and reproducibility.
• Compare and evaluate multiple LLM families (GPT/Gemini and others).
• Translate vague product requirements into structured generation logic.
• Implement generation APIs with clear input/output schemas.
• Build scalable, production-ready pipelines for LLM inference.
• Implement robust error handling: retries, fallbacks, timeout strategies.
• Optimize latency, throughput, and overall system reliability.
• Implement routing between different models without affecting frontend behavior.
• Collaborate with backend engineers to design scalable service architecture.
• Ensure observability: logging, monitoring, tracing for model-based workflows.

Output Quality, Ranking & Evaluation

• Implement rule-based and embedding-based ranking logic.
• Define and maintain quality metrics with PM & AI Lead.
• Use user behavior signals (CTR, engagement, selections) to improve generation quality.
• Establish continuous feedback loops for prompt refinement.
• Maintain long-term quality stability as user diversity increases.

User DNA / Personalization Engine

• Analyze user-provided content and extract style/tone/patterns.
• Build embedding-based user style profiles.
• Implement personalization constraints into prompts.
• Build prompt augmentation mechanisms to strengthen personalized output.
• Collaborate with data engineering on storage structures, embedding tables, and scalable infrastructure.

Trend & Topic Intelligence (Nice to Have)

• Keyword extraction, clustering, and topic grouping.
• Build lightweight trend scoring models.
• Support Topic/Trend Wall pipelines using embeddings and similarity search.
• Implement rate limits, quotas, concurrency protections, and usage caps.
• Design model access tiers (Free / Pro / Studio) with permission logic.
• Prevent misuse or overuse of expensive models.
• Monitor API consumption and enforce cost-efficient routing.

Requirements (Must Have)

• Strong, hands-on experience in prompt engineering.
• Deep understanding of tone control, style consistency, and output stability.
• Experience building LLM workflows and generation APIs in production.
• Familiarity with embeddings, similarity search, and basic clustering.
• Ability to convert ambiguous product needs into structured generation logic.
• Ability to operationalize AE logic into functional, scalable services.
• Strong quality sense and attention to detail.
• Fast iteration speed with strong ownership mindset.

Nice to Have

• Python (FastAPI or Flask) experience.
• Experience implementing A/B testing pipelines.
• Familiarity with short-form content culture (YouTube/TikTok titles).
• Experience in lightweight ML or data modeling.
• Startup experience (fast moving, ambiguity-friendly).