E2E Networks Q2 FY27 Earnings Call — Analysis (NSE: E2E)
E2E Networks delivered revenue of ₹223.8 Cr (+411% YoY, +43% QoQ) with 76.6% EBITDA margins, driven by full utilization of its Blackwell cluster under a ~₹1,000 Cr multi-year Sovereign AI contract.
Result quality: stable — Steady quarter. Management sentiment: optimistic.
The take
Q2FY27 Revenue ₹223.8 Cr ( +411% YoY ) . New guidance — Q3FY27 next lot of 1,024 b200 gpus dep… 1,024 B200 . New story: Sovereign AI and open-source advantage .
Results
Revenue reached ₹223.8 Cr (+411% YoY, +43% QoQ) with EBITDA of ₹171.5 Cr (76.6% margin, +141 bps QoQ) and PAT of ₹72.7 Cr (32.5% margin).
Financial highlights
| Metric | Value | Change | Basis |
|---|---|---|---|
| Revenue | ₹223.8 Cr | +411% | yoy · Q2FY27 · Q2FY26 |
| Revenue | ₹223.8 Cr | +43% | sequential · Q2FY27 · Q1FY27 |
| EBITDA | ₹171.5 Cr | +45% | sequential · Q2FY27 · Q1FY27 |
| EBITDA Margin | 76.6% | +141 bps | sequential · Q2FY27 · Q1FY27 |
| PBT | ₹97.1 Cr | +66% | sequential · Q2FY27 · Q1FY27 |
| PAT | ₹72.7 Cr | +66% | sequential · Q2FY27 · Q1FY27 |
| PAT Margin | 32.5% | +450 bps | sequential · Q2FY27 · 28.0% in Q1FY27 |
| Diluted EPS | ₹3.48 | +65.7% | sequential · Q2FY27 · ₹2.10 in Q1FY27 |
| Revenue | ₹380.5 Cr | +376% | yoy · H1FY27 · H1FY26 |
| EBITDA | ₹289.4 Cr | +2.3x full FY26 EBITDA | none · H1FY27 · 76% margin |
| PAT | ₹116.6 Cr | yoy · H1FY27 · Loss of ₹16.3 Cr in H1FY26 |
Guidance
Management targets maintaining a 70%+ EBITDA margin in the medium to long term and plans to deploy the next tranche of 1,024+ B200 GPUs in the current quarter.
What management committed to
- We expect [the next lot of 1,024 B200 GPUs] to go live this quarter [Q3FY27]. — 1,024 B200, Q3FY27
Key themes
GPU monetization and sovereign AI expansion
How the narrative shifted
- Sovereign AI and open-source advantage: Management emphasizes the Sovereign AI platform (TIR and Jarvis) as the key to future growth, enabling full-stack operation and global adoption.
- GPU capacity expansion super cycle: Management continues to position AI infrastructure build-out as a massive, long-term super cycle with a global trillion-dollar TAM, and sees no near-term competition concerns.
- Operating leverage and margin sustainability: Management downplayed the 88% gross margin as not a target, refocusing on EBITDA margin.
- Customer migration to longer-term contracts: A nearly INR1,000 crore multiyear contract (through June 2029) was signed, converting existing pay-as-you-go Blackwell capacity to long-term committed revenue, validating the migration thesis.
- International expansion and alliance management: International revenue was cited at 20-25% as a point-in-time data point, but management minimized its trackability and reiterated the arm's-length relationship with L&T and openness to partnerships.
- Input cost pressure on CPU hardware: No mention of CPU hardware cost pressures in the current call; the topic has fallen out of the narrative.
- Platform-led differentiation and product development: Management highlights heavy investment in the Sovereign AI platform (TIR, Jarvis), using own GPUs for R&D, and soft-launching inference products like Blaze and Jarvis Lab, aiming for long-term technology adoption.
- Global AI infrastructure demand supercycle and competitive benignity: Management repeatedly argues that the global TAM for AI infrastructure is enormous (~$1 trillion), leaving ample room for all players, and downplays competitive intensity as a near-term concern.
Operational commentary
- Signed a nearly ₹1,000 Cr contract with a Sovereign AI company running through June 2029, converting existing Blackwell B200 GPU capacity from pay-as-you-go to a multi-year committed structure.
- The next planned batch of 1,024+ B200 GPUs is scheduled to go live during the current quarter (Q3FY27).
- India AI Mission customers accounted for approximately 13% to 15% of Q2FY27 revenue.
- International customers contributed approximately 20% to 25% of Q2FY27 revenue.
- Strengthened leadership and advisory talent with the appointment of a new Chief Revenue Officer, Field CTO, Founder's Office additions, and Anandamoy Roychowdhary as Strategic Advisor.
- Soft-released new inference products including Blaze Endpoint and updates to Jarvislabs, transitioning customer workloads from pure compute rentals to the proprietary Sovereign AI software stack (TIR and Jarvis).
Analyst Q&A
Q. Is the MRR completely baked in from the ₹1,000 Cr contract, or is there incremental MRR to come?
Declined to quantify, stating exact contribution depends on usage over time and refuses to predict future MRR.
Q. Will gross margins remain sustainable at 88% and what is the outlook on absolute opex going forward?
Reframed the business target as 70%+ EBITDA margins for medium and long term, noting quarterly gross margins fluctuate based on the timing of capex and idle capacity before going live.
Q. What is the GPU utilization level and fleet realization across the platform?
Explained that point-in-time utilization is not the tracked metric because GPUs are heavily utilized for internal R&D platform development and customer usage patterns vary widely; fleet realization is the preferred metric.
Q. Can we assume 2,000 GPUs per annum as a base case for deployments over the next 3 years?
Management declined to endorse any forward-looking volume estimates, maintaining that capex updates will be provided only upon plan solidification.
Q. Are you exploring off-balance-sheet financing or partnering on idle capacity with players like L&T?
Confirmed they maintain an arm's-length partner relationship with L&T and remain open to all financing and distribution arrangements without committing to specific structures.
Research and educational content only. Not investment advice.