2 days ago

LLM & Agent Algorithm Expert - TikTok Search

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 8 hours agoStatus: Live

Job Summary

Build and iterate the search-domain LLM foundation and agent execution framework that powers TikTok Search’s shift from retrieval to proactive, long-horizon task agents. Lead the training–inference–evaluation loop with long-term memory, self-improvement (RSI), and reward/verifier systems. Develop high-quality vertical data synthesis and multimodal capabilities (image/video, visual search, OCR, TinyLLM), while exploring next-gen architectures and safety/alignment for large multi-agent systems.

Location: Singapore

Workplace: Onsite

Employment Type: Full time

Job Function: Administration & Executive Assistance

Key Responsibilities

•Lead R&D and iteration of the search-domain LLM foundation, including pre-training/post-training pipelines and inference optimization for long context and efficiency.
•Own the 0→1/1→N build-out of the agent execution framework (Harness), including orchestration of tool calling, planning, memory, environment interaction, and multi-agent collaboration/scheduling.
•Drive the agent loop and training–inference–evaluation closed loop by building long-term memory, causal reasoning, and recursive self-improvement (RSI) mechanisms.
•Build vertical-domain data synthesis and quality control, and develop online reward/verifier and benchmark evaluation for long-cycle, cross-domain tasks.
•Support business scaling by enabling long-horizon task agents and multimodal experiences (multimodal AIGC, visual search, on-device intelligence via object detection/OCR/TinyLLM).

Key Requirements

•Bachelor’s degree or above in Computer Science (or related) with AI algorithm R&D experience.
•Hands-on research and implementation experience in LLM/agent/multimodal/search, able to independently own and lead a technical direction.
•Strong mastery of an LLM/agent stack, including full-pipeline LLM (pre-training/SFT/RLHF/alignment/inference optimization/data synthesis) and/or full-stack agent systems (Harness, multi-agent scheduling, memory, agent loop, RSI, reward/verifier).
•Strong engineering for distributed training, mixed-precision training, and inference acceleration (quantization/pruning/distillation/TensorRT) plus large-scale data processing (e.g., MapReduce/Spark).
•Demonstrated familiarity with CV & multimodal and/or NLP foundations (e.g., retrieval/recognition, segmentation, object detection, OCR, multilingual learning, NLG).

Experience:AILLMAI agentsMultimodalSearchComputer visionNLP

Education:Bachelor's

Skills:Technical leadershipCross-team collaborationCommunicationResearch ownershipEngineering judgment

Tech Stack:LLMAgentsHarnessMulti-agent schedulingTool callingPlanningMemory mechanismsAgent LoopSelf-improvement (RSI)Reward/VerifierQuantizationPruningDistillationInference accelerationTensorRTMapReduceSparkSFTRLHFAlignment

Company Brief

TikTok

Short-form video platform that lets users create, share, and discover entertainment content through algorithmic recommendations. It also offers advertising and creator tools for brands, influencers, and businesses.

Industry: Digital Media

Company Size: Enterprise (1,001+ employees)

Growth: Scaleup

Headquarters: Singapore, Singapore

Founded: 2016

Website LinkedIn

Loading job details...

Preparing the role view and application actions.

TikTok

2 days ago

LLM & Agent Algorithm Expert - TikTok Search

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 8 hours agoStatus: Live

Job Summary

Location: Singapore

Workplace: Onsite

Employment Type: Full time

Job Function: Administration & Executive Assistance

Key Responsibilities

•Lead R&D and iteration of the search-domain LLM foundation, including pre-training/post-training pipelines and inference optimization for long context and efficiency.
•Own the 0→1/1→N build-out of the agent execution framework (Harness), including orchestration of tool calling, planning, memory, environment interaction, and multi-agent collaboration/scheduling.
•Drive the agent loop and training–inference–evaluation closed loop by building long-term memory, causal reasoning, and recursive self-improvement (RSI) mechanisms.
•Build vertical-domain data synthesis and quality control, and develop online reward/verifier and benchmark evaluation for long-cycle, cross-domain tasks.
•Support business scaling by enabling long-horizon task agents and multimodal experiences (multimodal AIGC, visual search, on-device intelligence via object detection/OCR/TinyLLM).

Key Requirements

•Bachelor’s degree or above in Computer Science (or related) with AI algorithm R&D experience.
•Hands-on research and implementation experience in LLM/agent/multimodal/search, able to independently own and lead a technical direction.
•Strong mastery of an LLM/agent stack, including full-pipeline LLM (pre-training/SFT/RLHF/alignment/inference optimization/data synthesis) and/or full-stack agent systems (Harness, multi-agent scheduling, memory, agent loop, RSI, reward/verifier).
•Strong engineering for distributed training, mixed-precision training, and inference acceleration (quantization/pruning/distillation/TensorRT) plus large-scale data processing (e.g., MapReduce/Spark).
•Demonstrated familiarity with CV & multimodal and/or NLP foundations (e.g., retrieval/recognition, segmentation, object detection, OCR, multilingual learning, NLG).

Experience:AILLMAI agentsMultimodalSearchComputer visionNLP

Education:Bachelor's

Skills:Technical leadershipCross-team collaborationCommunicationResearch ownershipEngineering judgment

Company Brief

TikTok

Industry: Digital Media

Company Size: Enterprise (1,001+ employees)

Growth: Scaleup

Headquarters: Singapore, Singapore

Founded: 2016

Website LinkedIn