Tara Madhyastha

Who they are

Tara Madhyastha is Founding Enterprise Account Executive at basetwo.ai — a research scientist and solutions architect who spoke at USENIX FAST in 2003 and crossed from HPC/cloud infrastructure into enterprise AI sales.

Person

Tara's career began in research — she holds a Ph.D. and was a Principal Research Scientist, with an early public footprint that includes organizing USENIX FAST in 2003. She later complemented that technical base with a Mini-MBA for Engineers and Technology Managers from Rutgers University–New Brunswick, a credential that signals a deliberate pivot toward the commercial side of deep tech. From research she moved into solutions architecture: stints at Rescale (HPC cloud for simulation workloads), together.ai (large-model infrastructure), and CoreWeave (GPU cloud) trace a consistent focus on compute-intensive, technically demanding environments. She also spent time at Amazon Web Services in a public-sector capacity, publishing on the AWS blog. The through-line is deep infrastructure expertise — HPC, cloud, AI compute — progressively repositioned toward customer-facing work in enterprise and industrial verticals. She has spoken at PEARC, ICLR, AIChE, and the CoreWeave Fully Connected Conference, with themes spanning research computing, HPC/cloud/AI, and manufacturing-adjacent technical audiences. Possibly — her content interests in Life Sciences, Chemicals, and Process Optimization reflect the industries she now sells into at basetwo.ai.

Company

Basetwo's most recent move is a November 2025 strategic distribution partnership with NS Solutions Corporation in Japan, extending its reach into the Japanese industrial market. That follows the January 2025 Series A of $11.5 million CAD led by AXA Venture Partners, with a notably industrial investor syndicate — Shimadzu Corporation, Chiyoda Corporation, Glasswing Ventures, Deloitte Ventures, and Global Brain Ventures among others. The company's Physics AI platform targets pharmaceutical and chemical manufacturers, claiming up to 40% improvements in cycle times and raw material usage and 25% improvement in product quality. Basetwo is now building AutoPilot, an autonomous real-time manufacturing control layer, funded by the Series A. As of 2025 the team sits at roughly 20 people across Canada, the US, and the UAE, with active hiring planned in Toronto, Vancouver, and Calgary.

Market

Basetwo plays in manufacturing process optimization — specifically the overlap of chemical engineering simulation, physics-based modeling, and AI — targeting pharmaceutical and chemical producers who need measurable yield and quality gains. The company positions itself against process optimization incumbents; tracxn surfaces FIS, State Street, and SimCorp as named competitors, though those are financial-infrastructure players and may reflect dataset noise rather than true head-to-head rivals in industrial AI. The broader industry dynamic is one of increasing AI regulation and data-sovereignty pressure in 2025, which is pushing industrial AI vendors to localize and comply market-by-market — a dynamic Basetwo's Japan partnership signals it is navigating actively.

Network

No direct relationship edges are available for Tara. The most relevant named figure at basetwo.ai from the claims is CEO and Co-founder Thouheed Abdul Gaffoor, who leads the company's Physics AI vision. Tara's conference appearances at PEARC, ICLR, AIChE, and CoreWeave Fully Connected suggest a network that spans research computing, AI infrastructure, and industrial process communities.

Thouheed Abdul Gaffoor· CEO and Co-founder, basetwo.ai

How they likely show up

Ph.D. research background followed by a Mini-MBA → she bridges deep technical credibility with commercial framing; likely comfortable speaking both languages in the same meeting.
Career path through Rescale, together.ai, CoreWeave, and AWS — all compute-infrastructure companies — → she builds domain depth iteratively rather than jumping industries, and probably goes deep before going broad on any new account.
Founding Enterprise AE title at a ~20-person company → she is in build mode, not playbook-execution mode; expect high ownership and tolerance for ambiguity.
Speaking record at USENIX FAST, ICLR, PEARC, and AIChE → comfortable in expert-practitioner rooms; likely more at ease with engineers and scientists than with purely commercial buyers.
Possibly — content themes in Life Sciences, Chemicals, and Process Optimization suggest she has oriented her technical vocabulary toward the specific pain points of the industries she now sells into.

Conversation tips

→ Lead with the technical substance — she has a Ph.D. and spent years in research and solutions architecture; don't oversimplify the product or the problem.
→ Ask about her transition from Principal Research Scientist to enterprise sales — it was deliberate (the Rutgers Mini-MBA signals that) and she likely has a considered view on why it made sense.
→ Reference her HPC and cloud infrastructure background when discussing Basetwo's Physics AI platform — she almost certainly frames the product through that lens with technical customers.
→ Acknowledge the Japan partnership or the industrial investor syndicate (Shimadzu, Chiyoda) — showing you read the specifics signals you did the work.
→ Don't assume she's a pure commercial operator — she's a founding AE at an early-stage company, which means she's also shaping the sales motion itself.

Toolbox

Openers

Open on the NS Solutions Japan distribution deal announced in November 2025 — it's the freshest move and opens a natural conversation about how Basetwo is sequencing international expansion from a ~20-person base.
Reference her USENIX FAST 2003 appearance as speaker and organizer — it's an unusually early and specific public-record moment that shows you looked beyond LinkedIn, and it anchors a conversation about her long arc from systems research to industrial AI sales.
Bring up the investor syndicate on the Series A — Shimadzu and Chiyoda are operating industrial companies, not just financial VCs; that's an unusual signal about Basetwo's go-to-market strategy and worth asking about.

Discovery questions

The Series A syndicate includes Shimadzu and Chiyoda as strategic investors — how does having operating industrials on the cap table change the conversations you have with prospective customers?
Basetwo's Physics AI platform claims 40% cycle-time improvements — what does the enterprise discovery process look like when the buyer is a process engineer rather than a CTO?
You moved from Principal Research Scientist through solutions architecture roles at Rescale, together.ai, and CoreWeave before joining Basetwo as a founding AE — what made enterprise sales the right next move after building that infrastructure background?

Avoid

Don't treat her as a conventional enterprise sales rep — she has a research Ph.D. and a decade in technical solutions roles, and oversimplifying the technology or the buyer problem will signal you haven't done your homework.

Make it yours