d-Matrix
Circuit mix
Where d-Matrix shows up, by circuit. Each event counts once: 2 if it sponsored, 1 if it only had speakers. Hover a segment for the conference.
- AI engineering & MLOps60%3 events
- Research20%1 event
- Software engineering & trading tech20%1 event
Presence by conference and year
Each cell is one conference in one year. Fill shows the company sponsored, shaded by its best tier that year; a dot means it had speakers there. Hover for details; click to open the event.
- Title
- Platinum
- Gold
- Silver
- Bronze
- Other / unlabeled tier
- Had speakers
| Conference | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | 2026 | 2027 | 2028 |
|---|---|---|---|---|---|---|---|---|---|---|
| PyTorch Conference | ||||||||||
| STAC Summit | ||||||||||
| NeurIPS (Conference on Neural Information Processing Systems) |
Activity by year
Three measures on one year axis; each panel has its own scale. 2027, 2028 is still mostly ahead, so its bars are faded and incomplete.
Conferences present at
Distinct conferences each year, split by whether the company sponsored or only had speakers there.
- Sponsored
- Speakers only
Distinct speakers
People from this company on a program that year, each counted once.
Sponsorships by tier
Sponsorship rows per year (one per event and listing), stacked by normalized tier.
- Gold
- Bronze
- Other / unlabeled
Where they've grown or pulled back
FlatSponsored 2 conferences in 2026, flat from 2 in 2024.
Counted per conference. 2026 is the latest year whose conferences have mostly happened. Tiers use the best tier each year; swatches show the path, oldest first. A conference counts as pulled out only when a later event published a complete sponsor list, since most events don’t say whether their list is complete.
Moved up 1
Best tier rose between the first and latest year with a ranked tier.
- PyTorch ConferenceBronze 2024 → Gold 2026
Once or on and off 1
No clear direction, or a gap we can’t confirm because later sponsor lists are incomplete.
- NeurIPS (Conference on Neural Information Processing Systems)once, 2024, Gold
Not on the company map: lookalikes are computed only for companies present at four or more conferences.
Often sponsors alongside
Companies that sponsored the same events, by number of shared events (out of 5 this company sponsored).
| Company | Shared events | Share |
|---|---|---|
| 5 | 100% | |
| IBM | 5 | 100% |
| Intel | 5 | 100% |
| Amazon Web Services | 4 | 80% |
| AMD | 4 | 80% |
| Jane Street | 4 | 80% |
| Lambda | 4 | 80% |
| Lightning AI | 4 | 80% |
| Meta | 4 | 80% |
| Anyscale | 3 | 60% |
| Arm | 3 | 60% |
| Baseten | 3 | 60% |
| Cloud Native Computing Foundation | 3 | 60% |
| The D. E. Shaw Group | 3 | 60% |
| Databricks | 3 | 60% |
Sponsorships 5
| Event | Date | Place | Tier as listed | Normalized tier |
|---|---|---|---|---|
| STAC Summit New York, Fall 2026 | Oct 21, 2026 | New York, United States | Tech discovery | Other |
| PyTorch Conference North America 2026 | Oct 20–21, 2026 | San Jose, United States | Gold | Gold |
| PyTorch Conference 2025 | Oct 22–23, 2025 | San Francisco, United States | Bronze | Bronze |
| NeurIPS 2024 | Dec 10–15, 2024 | Vancouver, Canada | Gold | Gold |
| PyTorch Conference 2024 | Sep 18–19, 2024 | San Francisco, United States | Bronze | Bronze |
Speakers 7
| Speaker | Event | Role | Talk | Links |
|---|---|---|---|---|
| Kam Eshghi | STAC Summit New York, Fall 2026 | speaker | Breaking the Memory Bottleneck in AI Inference | — |
| Vid Jain | STAC Summit New York, Fall 2026 | panelist | Closed vs open: Specialized AI systems for quant finance | — |
| Tristan Webb | PyTorch Conference 2025 | poster | Poster Presentations: Ecosystem & Performance | — |
| Xin Wang | PyTorch Conference 2025 | poster | Poster Presentations: Ecosystem & Performance | — |
| Zifei Xu | PyTorch Conference 2025 | poster | Poster Presentations: Ecosystem & Performance | — |
| Tristan Webb | PyTorch Conference 2024 | lightning talk | Lightning Talk: d-Matrix LLM Compression Flow Based on Torch.Fx: Simplifying PTQ/QAT | — |
| Zifei Xu | PyTorch Conference 2024 | lightning talk | Lightning Talk: d-Matrix LLM Compression Flow Based on Torch.Fx: Simplifying PTQ/QAT | — |