Early-career researcher at the intersection of LLM multi-agent research automation and ML-based compiler optimization
2
papers here
2
reviewed by Evolve
0
first-author
1
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-19) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
The indexed output record is too sparse to score: the OpenAlex identity shows an h-index of 0 and only a single 2026 work. This axis would measure cumulative publication volume, authorship positions, and career-length-normalized productivity across venues.
Impact—
No citation-based impact signal is available yet: h-index and i10-index both stand at 0, with essentially no citation window for the 2026 papers. This axis would measure citation uptake, influence on follow-up work, and recognition within the subfield.
Focus0/100
Focus scores 0/100: the raw topic data lists three single-paper topics (parallel computing and optimization techniques; cloud computing and resource management; logic, programming, and type systems), and fieldDist splits evenly at 0.5/0.5 between LLM multi-agent research automation and ML-based compiler optimization. With only 1 indexed work in three years, this spread reads as an exploratory, not-yet-consolidated agenda rather than established breadth.
Collaboration46/100
Collaboration scores 46/100. The record lists 9 distinct coauthors, including frequent collaborators Yifan Zhang, Yunheng Zou, Shaokun Zhang, Jian Hu, Binfeng Xu, Jan Kautz, Yi Dong, Mingqiao Mo, and Yunlong Tan. On the two reviewed papers Zhang is a middle author on Agora but last and corresponding author on CompileRover, suggesting a shift toward coordinating roles on small, arXiv-first projects.
Activity20/100
Activity scores 20/100: recent3yWorks is just 1, and countsByYear contains a single entry (1 work in 2026). Both reviewed papers are 2026 arXiv preprints, so visible activity concentrates in one recent year and may undercount output not captured by the OpenAlex identity.
Reproducibility38/100
Reproducibility scores 38/100, the mean over 2 reviewed papers. Agora rates medium: it builds on Git-based provenance and reports a 12-day, 13-agent run reaching 1.899 bpb, though the causal benefit over a plain log is untested. CompileRover rates low: the review cites pseudo-math metrics, wrong citations, internal contradictions, and a broken worked example. Verifiable artifacts are the clear next step.
Research summary
Hao Zhang is an early-stage researcher whose visible work sits at the intersection of large language model multi-agent systems and machine learning for systems, specifically compiler optimization; the two areas split the field distribution evenly at 0.5 and 0.5. No institution, country, or ORCID is attached, and because 'Hao Zhang' is a very common name, this profile is name-resolved from OpenAlex bibliographic data with limited identity confidence. The indexed record is thin: h-index 0, i10-index 0, one work in the last three years, and countsByYear shows a single entry of 1 work in 2026. Evolve has reviewed two 2026 arXiv papers. On Agora, which frames Git-committed research DAGs as shared memory for LLM agents and reports a 12-day, 13-agent run reaching 1.899 bpb, Zhang is a middle author; reproducibility readiness is rated medium, with the causal benefit over a plain log still untested. On CompileRover, a tri-role LLM framework for virtual machine compiler optimization, Zhang is last and corresponding author; the review flags pseudo-math metrics, wrong citations, internal contradictions, and a broken example, and rates reproducibility low. The reproducibility mean across the two reviewed papers is 38/100, with 9 distinct coauthors recorded.
Collaboration pattern
With only 9 distinct coauthors recorded, Zhang works in small, arXiv-first teams. The frequent-collaborator list — Yifan Zhang, Yunheng Zou, Shaokun Zhang, Jian Hu, Binfeng Xu, Jan Kautz, Yi Dong, Mingqiao Mo, Yunlong Tan — suggests overlapping pools across the two 2026 projects. The shift from middle author on Agora to last and corresponding author on CompileRover points toward a coordinating role, though the collaboration score of 46/100 stays modest given the sparse record.
Research keywords
Highlights
Agora: Git-committed research DAG as shared memory for a 13-agent auto-research run
CompileRover: last-and-corresponding authorship on a tri-role LLM compiler framework
Timeline
- 2026Agora arXiv preprint appears: a middle-author contribution proposing Git as shared memory for collective LLM auto-research, with a 12-day, 13-agent run reaching 1.899 bpb.
- 2026CompileRover arXiv preprint appears: last and corresponding author on a tri-role LLM framework for virtual machine compiler optimization.
- 2026The OpenAlex snapshot for this identity records 1 work in 2026, h-index 0, and 9 distinct coauthors — a publication record just beginning to accrue.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric counts (h-index 0, i10-index 0, 1 work in 2026) come from an OpenAlex identity match only; no ORCID or institutional link was available to confirm the record.
- 'Hao Zhang' is an extremely common name; this profile is name-resolved from public bibliographic data and may mix records or omit indexed work.
- Only the two Evolve-reviewed 2026 arXiv papers inform the qualitative assessment; other outputs, if any, are not covered.
- Institution and country are unknown; the focus score of 0 partly reflects data sparsity rather than confirmed absence of research concentration.