Frontier Engineering

Solving global tech challenges, sharing insights, and empowering developers

Latest posts

BERJAYA
Sep 24, 2026
Post likes count 5

Documentation Is the New Source Code

BERJAYA
Liza Shakury

A field guide to building a layered, AI-readable documentation system — instruction files, skills, ADRs, and tests — that turns your repo into a reliable substrate for Copilot, agents, and high-velocity engineering workflows.

BERJAYA
Sep 14, 2026
Post likes count 3

Building a Japanese LLM Evaluation Pipeline: Lessons from a Two-Day Hackathon

BERJAYA BERJAYA BERJAYA
Jaein,
Akihiro,
Kohei

ISE hackathon produced a reusable Japanese LLM evaluation pipeline with multi-turn, facet-based scoring, cross-provider model support, and publication-ready benchmarking outputs.

BERJAYA
Jul 17, 2026
Post likes count 0

Teaching a Vision Model to See Like a Human Annotator—and Catching It When It Lies

BERJAYA BERJAYA
Deeptanil,
Kartheek

A multimodal LLM enrichment pipeline that extracts structured metadata from visual assets, constrains output to predefined values to minimize hallucinations, and uses a ground truth evaluation template to measure quality—all as a plug-and-play module.

BERJAYA
Jul 9, 2026
Post likes count 0

From Noisy Queries to Precise Frames: Query Decomposition for Media Asset Search

BERJAYA BERJAYA
Kartheek,
Deeptanil

How query decomposition separates metadata filters from visual intent to significantly improve media asset retrieval quality.

BERJAYA
Jul 3, 2026
Post likes count 0

Enabling MLflow OpenAI Autolog on PySpark Workers

BERJAYA
Jaya Kumar

When distributing LLM calls across PySpark workers via mapInPandas, MLflow autolog silently fails. Here is how to fix it.

BERJAYA
Jun 26, 2026
Post likes count 1

Passing Context Between Agents in Multi-Agent A2A Systems

BERJAYA BERJAYA
Olha,
Scott

How we solved the challenge of passing conversational context between independently deployed agents using the A2A protocol's embedded context pattern, keeping domain agents stateless and secure.

BERJAYA
Jun 19, 2026
Post likes count 1

Separating Deterministic Extraction from AI Inference in Industrial Summarization

BERJAYA
Viktor Ciroski

When 100% of our prototype outputs were valid JSON but 0% met the data contract, we discovered the LLM was doing work that software should own. A field classification exercise and 4-pass pipeline brought schema compliance from 0% to 100%.

BERJAYA
Jun 12, 2026
Post likes count 3

Orchestration Patterns for Multi-Agent Systems: Performance and Trade-offs

BERJAYA
Lily Jia

Real-world lessons from evolving a production chatbot into a coordinator-based multi-agent architecture, including performance trade-offs for enterprise-scale agent reuse.

BERJAYA
Jun 4, 2026
Post likes count 4

How we Decide Between Keyword and Hybrid Search: 5 Enterprise Evaluation Criteria

BERJAYA
Beijie Zhang

A data-driven framework we use in enterprise deployments to decide between vector-only keyword and hybrid search, based on five measurable evaluation criteria.