The business unit inside Hitachi’s Digital Systems & Services (DSS) sector formed by merging Hitachi Digital Services and GlobalLogic; the combined organization is called Hitachi Digital. CEO: Anand Birje, Santa Clara, California. Tagline: “Engineer what the world depends on.” ⟨No source the wiki holds dates the renaming itself. What is dated: Birje’s appointment to the BU in August 2026 and to the combined organization in September 2026 — Q2 FY2026, Hitachi’s fiscal year running April–March — and the deck’s own “Hitachi Digital | Q2” footer. An earlier version of this line gave the rename as “the second half of FY2026”, which no source supports and which would put it after the September appointment; corrected by the 2026-11-02 lint.⟩
⟨Inference: this is the unit a SINE mechanism would most plausibly sit inside — see Why this page exists below.⟩
Status ⟨2026-11-01⟩
Two primary sources, both supplied by Tell Münzing: Anand Birje’s 28 October 2026 presentation (12 slides, “Hitachi Digital | Q2”) for the BU, and Jun Abe’s 10 June 2026 Investor Day deck (25 slides) for the sector above it. Everything below is from the Birje deck unless marked otherwise. Besides Birje and Abe, no person in the DSS digital organization is named in any source the wiki holds, and no contact exists.
Where it sits
- Sector: Digital Systems & Services. FY2025 revenue ¥3.1tn, adj. EBITA margin 15.5% — the highest-margin of the four sectors — Lumada revenue ratio 62% (Inspire 2027 deck, pp. 9–10; but see open task 20 on the segment restatement).
- Sector CEO: Jun Abe, EVP, whose 10 June 2026 Investor Day deck the wiki now holds. Birje reports into the sector, not at it. Sector FY2026 plan: revenue ¥3,190.0bn, margin 15.7%, Lumada ratio 70% rising to 75–80% by FY2027 (raised from ca. 65% a year earlier); backlog ¥1.8tn, +11%.
- The money: ¥30bn organic investment in FY2026, whose technology column opens with “enhance data fabric capabilities”, plus financial capacity for up to ¥1tn inorganic over FY2026–27 (Abe p. 18).
- The BU: “Inside Hitachi’s Digital Engineering & AI Solutions BU · Full-stack digital engineering with IT-OT · External clients plus sell-to / sell-with Hitachi units”.
⟨Background: GlobalLogic was acquired by Hitachi in 2021 for ≈$9.6bn; no
source in sources/ states this and the deck does not mention it.⟩
Anand Birje
Appointed CEO of the Digital Engineering & AI Solutions BU in August 2026 and CEO of the merged Hitachi Digital organization in September 2026 — both within the last three months of the deck’s date.
Prior: Encora, named on his own slide as a “GlobalLogic competitor”, CEO for three years through to its $2.35bn acquisition by Coforge in April 2026; HCLTech, 15+ years, most recently President of a $4.5bn Digital Business Services unit (digital consulting, app services, data engineering) and founding leader of an infrastructure and cloud business that reached $2.5bn+ within a decade; Emerson Electric and Capgemini — “industrial control systems, SCADA and MES, hands-on IT-OT convergence”. Stated mandate: unify GlobalLogic and Hitachi Digital Services capabilities while scaling digital transformation “across Global markets and mission critical industries leveraging AI, Data and Cloud”.
⟨Inference: an IT-services operator with a deal history and genuine OT background, three months into the job, consolidating two organizations and rebranding them. The deck’s own closing line in the Opportunities column — “Evolution need: Strong coherence of strategy, leadership and execution” — is a new CEO naming his inherited problem in front of an investor-cycle audience. ⟨Inference: a first approach in FY26 is arriving during a reorganisation, which cuts both ways — attention is scarce, but the architecture is still being decided and the HMAX CTO office does not yet exist.⟩⟩
What the business actually is
The horizontal under HMAX, not an HMAX product. The three HMAX products — Energy, Mobility, Industry — belong to the three physical sectors. What this BU owns is the stack beneath them (deck p. 4):
| Layer | Function |
|---|---|
| Physical AI FDE service | Frontline experts alongside customers, continuously capturing knowledge |
| HMAX AI Operations | ”Ensure stable and optimized AI operations while preventing inappropriate AI behavior” |
| HMAX Data Fabric | Data platform — “capture and manage frontline knowledge as knowledge assets that AI can utilize”; IWIM / AI — “make decisions based on frontline context and constraints using ontologies and physical models” |
| Existing IT systems | Cloud and on-premises |
Plus HARC (Hitachi Application Reliability Centers), the named IP investment: “Evolution of HARC for Industry specific ontology and context layer acceleration”.
The same architecture one level up, and it is not the same ⟨2026-06-10⟩. Jun Abe (p. 14) draws HMAX as “the OS that supports social infrastructure operations”, resting on four pillars:
| Abe’s pillar (sector) | Birje’s layer (BU) |
|---|---|
| Domain Knowledge × AI — Physical AI | Physical AI FDE service |
| Talent — Physical AI FDE Team | (same team, drawn as the top layer) |
| Security — Mission-Critical Security | — no counterpart — |
| Data Infrastructure — Data Fabric | HMAX Data Fabric + HMAX AI Operations |
⟨Inference, and it is a routing fact rather than a contradiction: the sector CEO’s HMAX has a security pillar and funds “expand OT security services”; the BU CEO’s stack has no security layer and no confidentiality, provenance or verification vocabulary at all. The vocabulary SINE needs exists at sector altitude and thins out one level down. Caveat: Abe’s “security” means operational safety of infrastructure that must never stop — OT security, not provenance between parties. It is adjacent, not identical. Cite it as the reason the Data Fabric needs provenance; do not claim the pillar.⟩
And it is explicitly DSS’s ⟨2026-06-10⟩. Abe p. 16 puts a red band labelled “DSS Sector” under a triangle of HMAX Energy / Mobility / Industry, carrying “Data Fabric (Data Infrastructure)” and its two stated jobs: “integration and structuring of IT/OT data” and “enable cross-sector and cross-industry data utilization”. Ownership of the layer is therefore not an inference any more.
Shape of the business: 4 growth service lines (Cloud, Data, Physical AI, Enterprise Platforms); 7 key verticals (Energy, Mobility, Telecom, Pharma & Healthcare, Banking and Financial Services, Auto, Connective Industries); top 200 customers; 15 named ecosystem partners across an IT stack whose integration row reads “APIs / MCP / Orchestration” — AWS, Boomi, Claude, Databricks, Google Cloud, Microsoft, MuleSoft, NVIDIA, OpenAI, Oracle, PTC, Salesforce, SAP, ServiceNow, Snowflake.
The growth split
| Rate | What it is | |
|---|---|---|
| Synergy business — sell-to / sell-with Hitachi front business units | +46% YoY ⟨FY2026/H1 actual⟩ | Internal |
| Top 100 accounts, 65% of the business | single-digit 2-year CAGR | External |
The full year, from the sector’s books ⟨2026-06-10⟩. Abe p. 4 gives GlobalLogic revenue in dollars:
| unit: $m | FY2024 | FY2025 | change |
|---|---|---|---|
| Stand-alone | 1,950 | 2,000 | +2.6% |
| Synergies (booked by other BUs & Group companies) | 1,450 | 2,500 | +73% |
| Total | 3,400 | 4,500 | +32% |
Stated challenge on the same slide: “transformation of stand-alone businesses”. ⟨Inference: the external business is flat. All of the growth is internal, and the sector CEO says so in the only hard currency figures either deck provides. Birje’s +46% is the half-year of the same fact, reported more modestly.⟩
Named work (p. 11): Walgreens (pharmacy management modernization) — external; Hitachi High-Tech (agentic AI for semiconductor maintenance), Hitachi Energy (Physical AI with NVIDIA on factory floors and critical infrastructure), Hitachi Rail (“the FDE model for Smart Operations — including Factory IQ and vehicle manufacturing optimization”) — all internal.
⟨Inference, and the finding that matters most for the engagement: the internal channel is where this business grows. Three of four reference cases are Hitachi companies, and the internal motion runs at 46% against single digits externally. An approach that starts inside Hitachi is not a concession to SINE’s lack of an external reference — it is the part of this BU’s business that is working.⟩
Path forward
FY26 transform (rebrand; land the vertical model, the service lines and the HMAX-focused CTO office; launch partnerships; align to One Hitachi) → FY27 scale (top-200 accounts, M&A, margin through delivery transformation) → FY30 “dominate Physical and Digital AI-led business transformation”, at best-in-class peer growth and margin, with rising revenue per head and IP-led revenue “a far larger share of mix”.
Why this page exists — the fit, stated carefully
⟨Inference: three things in this one source point the same way.
- The layer matches. An attestation over signed sensor data is a property of a datum, which makes it a Data Fabric component — the layer this BU owns — not a vertical solution owned by Energy, Mobility or CI.
- The economics match. “IP-led differentiation … how growth decouples from headcount” and “IP-led revenue a far larger share of mix” describe exactly the kind of thing a mechanism is, as against a consulting engagement.
- The channel matches. Internal-first deployment is this BU’s fastest-growing motion, which is the same shape as Customer Zero — a term the deck never uses while practising it.
And one thing points the other way: the deck contains no confidentiality, provenance, verification, security or standards vocabulary at all. The only trust word is “trusted AI”, about model behaviour. So the fit is structural, not demonstrated: nothing in this source shows the problem SINE solves being felt at this altitude. Treat the match as a hypothesis with a clear falsifier — if the HMAX CTO office, once staffed, has no data-provenance remit, the hypothesis is wrong.⟩
Where SINE attaches, and to whom ⟨2026-11-01⟩
Filed at Tell Münzing’s request after the 1 November reading of the Abe and Birje decks. Everything here is inference built on the two DSS sources plus the Rail sustainability material already in the wiki; it is a working position for conversations, not a sourced finding.
1. Which layer — Data Fabric, or AI Operations?
Written in the Data Fabric; argued from AI Operations. Both, but not symmetrically.
- The Data Fabric is the insert point. An attestation is a property of a datum, created at capture. That is the layer’s own stated job — “capture and manage frontline knowledge as knowledge assets that AI can utilize” (Birje p. 4), “integration and structuring of IT/OT data” (Abe p. 16). Same shelf as lineage and schema.
- AI Operations is the argument. “Preventing inappropriate AI behavior” (Birje p. 4) is about the model. But after an incident in a mission-critical setting the question is what did it know, and was that true — which cannot be answered from the model side. IWIM exists because of “large-scale power outages and railway accidents resulting from incorrect AI decisions”; provenance is a precondition for that audit, not a parallel concern.
- Do not claim the AI Operations box. It is crowded and partly vendor-held — NVIDIA, ServiceNow and the model vendors all sit on Birje’s partner slide. Claiming it makes SINE a competitor to a named partner in the first sentence. Claiming a provenance property in the Data Fabric that makes AI Operations auditable makes SINE a component.
- The Data Fabric is contested too, but less so. Databricks and Snowflake are on Birje’s 15-partner slide and lineage is their home ground. Two things cut the other way: Abe’s six sector-level partners are all compute or model layer — no data-layer vendor is named where the ¥30bn is allocated — and lineage answers a different question. Inside one company, lineage suffices. The moment a datum crosses a company boundary, lineage is a promise and a proof is a proof.
- A third attachment, offered as a hypothesis. The business model is “continuous knowledge circulation”: frontline rules and tacit knowledge flow from the customer into reusable assets Hitachi resells across seven verticals. That is Hitachi monetising a customer’s know-how. Every sophisticated customer eventually asks what leaves their site, and neither deck has an answer. A mechanism that proves a result without exporting the data is then not a compliance feature but an unblocker for the engine — the version of the argument a P&L owner would hear.
2. The nuance for Hitachi Rail’s CSO — circular economy and value-chain data
The DSS decks and Dr. Shibasaki’s agenda are not the same problem, and the briefing should not pretend otherwise. Where they touch, precisely:
- Hitachi’s own first example is a value-chain coupling. Abe p. 16’s leading Data Fabric use case is “optimal production linked to green power supply and demand”. Energy data governing production decisions is the same shape as material and emissions data governing design and procurement decisions. The pattern is already drawn; only the domain differs.
- But every arrow on that slide is Hitachi-to-Hitachi. Energy ↔ Industry, Energy ↔ Mobility, Industry ↔ Mobility. No arrow reaches a supplier, an operator, a customer or an assurer. “Cross-industry” means across Hitachi’s own industries.
- The CSO’s problem lives exactly where the arrows stop. Rail reports actual emissions data from 76 suppliers, 14.93% of Scope 3 Category 1; roughly 85% of the purchased-goods footprint is estimated, not measured; 428 of 700 group procurement partners have reduction targets. The binding constraint is upstream and outside the company — see the consolidated question set, II1. That is one step beyond what the Data Fabric is drawn to do.
- Same for the asset side. HMAX Mobility serves 2,500 trainsets, over 35% built by another OEM; “Enhanced by HMAX” instruments every new Hitachi vehicle from 2027. Data Hitachi captures itself is a Data Fabric problem that Hitachi has already solved. Data Hitachi must be given is not drawn anywhere in either deck.
- The one-sentence form of the gap, unchanged: Hitachi can integrate data it is given; for data it is not given, it has no architecture.
- The nuance that keeps this honest. Nothing in the DSS material says Hitachi feels this as a problem. Abe’s stated risks are an IT-investment slowdown and component prices. The sustainability urgency is real in the CSO’s organisation and invisible in the sector’s. Two consequences: a briefing must carry the CE case to DSS rather than assume DSS is looking for it; and the CSO’s own demand signal — a tender requirement, an ISO 14067 boundary decision, a supplier that refuses — is the evidence DSS does not have and would need.
3. The narrative when the counterpart does not exist yet
The HMAX-focused CTO office is an FY26 deliverable on Birje’s own roadmap. It is unstaffed. That is the opportunity, and it is also why no name can be produced. Three beats, each repeatable internally without anyone having to own it:
- The decision is already made. Frontline data becomes a knowledge asset AI acts on — FDE, Data Fabric, IWIM. Nobody internally is arguing.
- One property of that data has no owner. Whether a figure derived from it can be checked by someone not allowed to see it. Today that is carried by contracts and NDAs, which work inside Hitachi and stop at the company boundary.
- The office that would own it is being built this fiscal year, and the budget line exists — Abe’s ¥30bn FY2026, technology bullet one, “enhance data fabric capabilities”.
The ask is a question, not a pitch: who is drawing the HMAX Data Fabric architecture, and is anyone responsible for data provenance in it? Cheap to answer, only an insider can answer it, and the answer is a name.
4. The internal audience map
Not equivalent, and six is too many to open at once — the useful request to the CSO is to rank them.
| Audience | Evidence it exists | What it is for |
|---|---|---|
| HMAX-focused CTO office (Hitachi Digital / DSS) | Birje path-forward, FY26 | The eventual owner. Unstaffed — target, not door. |
| Physical AI FDE Team | Abe p. 15: GlobalLogic + Hitachi Digital Services (AI, security, data infrastructure) meeting OT-sector domain experts; already inside Rail on Factory IQ (Birje p. 11) | The crossing point between DSS and the sectors, and the cheapest first conversation. A practitioner reference travels further here than an executive introduction. |
| CI Customer Zero programme | 175 in-house manufacturing sites (Abe p. 13) | The site, if a run is needed. Different sector, different budget. |
| Energy / EcoSpace | Hitachi § EcoSpace | The only unit already shipping a sustainability-data product. Nearest analogue — ally or incumbent, unknown which. |
| Group Standardization Committee / H-AIST CECRL | Hitachi § standardization apparatus · H-AIST CECRL | A separate door: Japan, slow, durable, standards-shaped. |
| Hitachi Rail CSO organisation | the engagement itself | The sponsor who can route. Not the buyer. |
5. Becoming legible to the DSS CEO and CTO
Legibility means using the units their own decks measure themselves in. The scoreboard, from Abe and Birje: AI-related revenue ¥800bn at CAGR 20–25%; HMAX ¥300bn at CAGR 50–60%; Lumada ratio 70% → 75–80%; adj. EBITA margin 16–18%; the gap to best-in-class multiples; IP-led revenue as a share of mix and “how growth decouples from headcount”; the synergy channel at +73% against a flat external book.
Six translations:
- IP, not a project. A reusable component shipping inside the Data Fabric, identical in every vertical — Birje’s “IP-led differentiation” verbatim. Never a study, a pilot or governance work; all three read as headcount, which is the thing their margin story is trying to shed.
- It opens the channel that is already growing. First run inside Hitachi, where 100% of the growth is; the artefact is resellable to the top-200 accounts. Internal-first is their fastest motion, not SINE’s fallback.
- A falsifiable test, not a programme. One figure from one Hitachi site, checkable by a third party, raw data never leaving the site. Pass or fail in weeks.
- Say what SINE is not, in one sentence, early: not a data platform, not a model vendor, not a competitor to anyone on the partner slide. Business leaders triage by threat before opportunity.
- Pick one of the seven verticals. Being inside a vertical they already fund is most of legibility. Energy and Auto are where cross-boundary supplier data is already a priced problem.
- Arrive protocol-shaped. Birje’s integration row reads “APIs / MCP / Orchestration”. An open mechanism that travels with a datum fits that row without needing a partner tier.
Format matters as much as content: a one-page component specification a CTO office can paste into an architecture diagram, not a memo. And the window is real — while the architecture is being drawn, a mechanism is a design input; once it lands, it is a change request.
Open questions
- Who will run the HMAX-focused CTO office, and does its remit include data provenance? Not answerable from any source held.
- How does this BU’s P&L relate to the DSS sector’s reported Lumada ratio of 62%? See open task 20.
- What is the BU’s own revenue? Not stated anywhere in the deck — the only figures given are growth rates and shares.
- Who is drawing the HMAX Data Fabric architecture, and is anyone responsible for data provenance in it? ⟨2026-11-01⟩ The cheapest question to ask internally; only an insider can answer; the answer is a name.
- Is “enhance data fabric capabilities” (Abe p. 18, technology bullet one of the ¥30bn FY2026 organic investment) this BU’s line, or the sector CEO’s? The budget is stated at sector level; the capability is stated at BU level.
- Does the Data Fabric have any cross-company arrow planned? All three value examples on Abe p. 16 are Hitachi-to-Hitachi. Whether a supplier-to-Hitachi or Hitachi-to-assurer path is out of scope or simply not yet drawn is the single most load-bearing unknown for SINE.
- Does the Security pillar (Abe p. 14) cover provenance? As stated it means OT operational safety — “infrastructure that must never stop” — not provenance between parties. Adjacent, not identical.
Cross-references
- Hitachi, Ltd.
- Anand Birje’s presentation, 28 Oct 2026 — the source
- DSS Business Strategy — Jun Abe, 10 Jun 2026 — the sector level above this BU
- Inspire 2027 progress deck
- Mobility
- Customer Zero
- Confidentiality gap in carbon accounting
Sources
sources/hitachi/2026-10-28-hitachi-digital-anand-birje-dss.pdfsources/hitachi/2026-06-10-hitachi-dss-sector-briefing-abe.pdfsources/hitachi/2026-04-27-hitachi-inspire-2027-progress-financial-analysts.pdf(sector figures)