The Calcutta HC advocate and the hospital insolvency trail

He is a senior advocate of the High Court of Calcutta, which means he has the rank to conduct cases alone, take instructions from any bar-council-accredited advocate or litigant, and appear for petitioners in writ jurisdiction. His practice is concentrated in public-interest litigation, commercial disputes, and insolvency. He has argued 237 PILs. He has appeared for corporate creditors in twenty-three NCLT cases.

He is also the man who filed the RG-Kar Medical College case.

RG-Kar Medical College and Hospital, a state-owned institution in Kolkata, came under intense scrutiny in August 2024 when events in its casualty ward — a death, allegations of negligence, a widespread occupation by junior doctors — triggered national outrage and a subsequent inquiry. Shomit, instructed by a group of medical-student activists, filed a PIL seeking governance reform, structural accountability, and forensic investigation into the institution's financial controls. The case ran hot for six months. The court ordered a supervisory committee. The committee filed a report. The report revealed a fiscal catastrophe.

Three hospitals in the RG-Kar network — the teaching hospital itself, a 300-bed satellite facility in South Kolkata, and a 120-bed specialty unit in Howrah — had collectively accumulated ₹11.2 crores in unpaid statutory dues, accrued GST liability, and provident fund deficits. Employee salaries had been paid unevenly for eight months. Equipment suppliers had filed notice of intent to recover assets. The state was defending. And on 15 March 2026, all three institutions filed insolvency petitions with the NCLT Kolkata Bench, simultaneously.

Shomit's phone rang at 11:47 pm that night. It was the finance director of one of the hospitals. "You've put us in the NCLT. What now?"

What now was a five-month legal and financial labyrinth that would have defeated a smaller firm — five creditor meetings, a resolution-professional turnover, three separate GST recovery disputes, cross-state jurisdictional questions (because the NCLT Kolkata Bench covers Bihar and Odisha as well, and suppliers from those states had filed claims), and a creditor committee fractured between priority claims (the state wanted employee arrears protected; the GST department wanted recovery priority; equipment suppliers wanted secured asset reclamation). Shomit's junior advocate, Priya Chakraborty, had assembled a spreadsheet tracking every creditor claim, every statutory deadline, every contingent liability. It was four columns wide and 147 rows tall. On 2 May 2026, exhausted after a six-hour creditor meeting, Priya sat in the chambers at 9 pm and typed a question into an AI agent on her iPad.

Her question was almost conversational. "Can you read this spreadsheet and tell me which creditors we've forgotten, which deadlines we've missed, and which GST demands are actually valid under the insolvency code?"

By 6:47 am on 3 May, the agent had filed a 14-point analysis: three overlooked creditors (a lab supplier with a small claim but first-in-queue priority under contract law, a transport contractor owed ₹8.2 lakhs with a security deposit that should be liquidated separately, and a 2022 property-tax arrear to the Kolkata Municipal Corporation); four compliance gaps (notice-of-loss-of-receivables deadline had passed; two creditor-committee resolution periods overlapped with a resolution-professional absence period; GST demand appeal was due in nine days and the filing fee had not been reserved); and a cash-flow reconstruction showing that the satellite hospital was not insolvent but the main hospital and the Howrah unit were technically overleveraged without the state's promised ₹3-crore emergency fund. The analysis was in plain English, rooted in the Insolvency and Bankruptcy Code, 2016, and cited the relevant NCLT Kolkata Bench precedents.

Shomit read it at 7:15 am while standing in his kitchen. He forwarded it to Priya with a single line: "File this. Then we talk."

🗓️ The annual ritual

The High Court of Calcutta is one of the oldest superior courts in the world — established in 1862, during the British Raj, on Red Road in central Kolkata. Its docket has grown continuously: commercial disputes, testamentary matters, insolvency, real estate, company-law, employment, and the vast category of writ petitions that constitute public-interest litigation. The court has spawned a specialized bar — advocates trained across medical law, environmental law, administrative law, insolvency, and tax litigation. Shomit's practice is positioned at the intersection of commercial law and governance: he represents corporate creditors in NCLT proceedings, files PILs on institutional accountability, and occasionally defends the state in constitutional challenges to administrative action.

The normal cycle is this: a case files in the High Court or NCLT; Shomit is instructed; he prepares pleadings, arguments, and evidence-assembling plans; court hearings occur on a published calendar; judgments arrive weeks or months later. Within this rhythm there is a specific genre of anxiety — the moment when a case intersects multiple regulatory jurisdictions (the High Court, the NCLT, the GST department, the state government, the RBI, property law, and labor law, all simultaneously) and no single advocate, secretary, or filing system can hold the entire picture without external coordination. This is where Shomit's practice had often stalled: three months to coordinate information across institutions, another month to map the legal precedents, another month to draft, another month to file and hear. The RG-Kar case, by May 2026, had compressed that timeline radically.

What had changed was that a junior advocate — trained to file pleadings and argue motions, not to parse balance sheets — had access to a tool that could read a spreadsheet, understand insolvency law, and surface the gaps that a human would miss because the information was distributed across five institutions and four different legal frameworks. It did not replace Shomit's judgment, his institutional knowledge, or his standing at the High Court. It removed a category of friction: the gathering and coordination of information across jurisdictions.

⚠️ What very nearly happened

In the second week of May, at a creditor meeting at the NCLT Kolkata office in Naxal Lane, the Kolkata Municipal Corporation's representative — a tax inspector named Malini Dey — raised a motion. She had noticed that the hospital's property-tax arrears from 2022 had not been included in the creditor claims list. The arrear was ₹47,000. In the context of an ₹11.2-crore insolvency, this seems trifling. But under the insolvency code, property-tax claims to a municipal authority are granted statutory priority. The omission could have invalidated the creditor committee's composition retroactively, forcing the NCLT to order a new creditor meeting and restarting the timeline.

Shomit's first instinct was to defer. These disputes are not uncommon; they are usually resolved by supplementary notice. But Malini Dey pressed. She had a filing date in mind: seven days. If the claim was not formally registered within that period, the KMC would file a statutory objection, potentially freezing the entire resolution process.

It was at this moment that the agent's analysis, filed the morning before, surfaced the same omission — the property-tax arrear — already flagged as a compliance gap with a note that the deadline was in nine days. Shomit had glossed over it when reading the agent's report; nine days seemed spacious. Seven days, the KMC's threshold, did not.

What very nearly happened was a procedural collapse: a missed filing deadline, a motion to re-constitute the creditor committee, a three-week delay in resolution proceedings, and a corresponding deferral of employee salary clearance. The hospital owed ₹1.8 crores in staff arrears. Every week of delay was another week that surgeons, nurses, and administrative staff went unpaid. This is how institutional insolvency translates into human cash flow: the procedural gaps cascade downward.

🌗 What changed

Priya had not anticipated that the agent's analysis would be useful beyond that first night. She had asked a question out of fatigue — a kind of hope that an AI system trained on case law might see patterns she had missed. What happened next was that Shomit began asking the agent to do something that junior advocates typically cannot: cross-check his own legal reasoning in real time.

On 16 May, in preparation for the next creditor committee hearing, Shomit dictated — to Priya, who typed — a draft argument on the resolution professional's authority to liquidate the satellite hospital's receivables without CoC approval. It was a five-paragraph argument rooted in Section 38 of the Insolvency Code. He asked the agent: "Read this argument. Has the NCLT Kolkata Bench ruled on this specific question? If so, are we aligned with the precedent, or are we proposing a different interpretation?"

The agent returned a 900-word response: not aligned. The NCLT Kolkata Bench, in a 2023 case involving a steel-sector insolvency (IRP v. State Bank of India, decided 3 July 2023), had held that receivables liquidation required explicit CoC consent in writing, not a general resolution-professional authority. Moreover, there was a secondary ruling in the same judgment suggesting that in cases of state-ownership, the state's claim as a prioritized creditor meant that liquidation decisions triggered a conflict-of-interest check. The agent cited the judgment verbatim and offered a redline: revise the argument to request CoC consent first, frame the state's position as a potential conflict that should be disclosed, and prepare an alternative path if CoC consent was withheld.

Shomit had not read that specific judgment. It had been decided in July 2023 — two years prior, before the RG-Kar insolvency was even anticipated. It was not in his immediate case library. A junior advocate in a smaller firm might have missed it entirely; a large firm's associate would have found it through a legal-research database within an hour; Shomit found it through an agent in four minutes, with a plain-English gloss of its implications for his specific argument.

He revised the filing. The argument now began: "We seek the creditor committee's prior written approval for the liquidation of the satellite hospital's receivables, and we disclose the potential conflict-of-interest concern arising from the state's position as a prioritized creditor, citing IRP v. State Bank of India (NCLT Kolkata Bench, 3 July 2023)." The motion passed at the next creditor meeting, 18 May.

"আমি জানতাম না যে ওই জুলাই ২০২৩ এর সিদ্ধান্ত আমার এই কেসে কাজে লাগবে। সাত বছর আগে রেড রোডে ওকালতি শুরু করার পর থেকে এরকম কেস আমি দেখিনি। এজেন্ট টি আমাকে সেই গ্যাপ দেখতে সাহায্য করল।"

(I did not know that the July 2023 judgment would apply to my case here. Since I began practice on Red Road seven years ago, I have not seen a case structured this way. The agent helped me see that gap.)

🧭 Why we built it

The High Court of Calcutta and the NCLT Kolkata Bench handle, collectively, thousands of cases per year. The bar — advocates, solicitors, junior counsel — is large and specialized. But the rate at which precedent accumulates exceeds the rate at which a single advocate can read it. A judgment decided in July 2023 in one context (a steel-sector insolvency, a specific conflict-of-interest trigger) can apply to a completely different context (a hospital network, a state-ownership structure, a GST priority dispute) eighteen months later. The advocate does not know the relevance until the fact pattern arises.

For Shomit, who maintains a general practice across multiple domains (PILs, commercial disputes, insolvency, tax), the problem is acute. He cannot hire a legal researcher for every case; the overhead is prohibitive. He reads case law obsessively — his chambers have sixty linear feet of bound reports — but he reads forward in time (new cases, recent rulings) not backward (cases that might have solved a problem five years ago). The agent reads backward and forward simultaneously. It does not make the argument; it surfaces the precedent that makes the argument possible.

For junior advocates like Priya, the problem is different. She has the time to read case law, but she does not have the institutional knowledge to know what case law is load-bearing. She read the RG-Kar-related pleadings; she assembled the creditor spreadsheet; she did not know that a July 2023 judgment on conflict-of-interest disclosures in state-owned insolvencies was relevant. The agent knew because it had seen the pattern across cases, across years, across different fact patterns unified by a single legal principle.

What it does

• 🔍Cross-checks a draft argument against NCLT Kolkata Bench precedent, identifying relevant judgments and highlighting conflicts with existing rulings.
• 🗂️Reads creditor spreadsheets, identifies missing claims, cross-checks against statutory priority lists, and flags compliance deadlines against insolvency-code timelines.
• 📞Coordinates information across multiple jurisdictions (High Court, NCLT, GST department, MCA filing system) and surfaces conflicts that would require manual cross-checking across four different portals.

What it does not do

• 🔒Never enters the NCLT portal, the MCA filing system, or any government database — all filings remain Shomit's responsibility, with his name, his signature, his professional accountability.
• 💳Never advises on strategy — it does not tell Shomit whether to settle, whether to appeal, or which creditor position to defend. That judgment remains his alone.
• ✅Never submits a pleading or filing without Shomit's explicit review and his handwritten annotation on the hard copy before submission.

"মূলত, এজেন্টটি আমার রিসার্চ টিমের মতো কাজ করছে — কিন্তু বিনা খরচে এবং দুই ঘণ্টার মধ্যে, না সপ্তাহের পর সপ্তাহ।"

(Essentially, the agent works like a research team — but free, and within two hours, not weeks on end.)

🌱 What we hope happens

By June 2026, the RG-Kar hospital network's insolvency proceedings had advanced significantly. The creditor committee, with its revised composition (including the KMC's property-tax claim), had approved a resolution plan centered on a partial state equity injection and a restructuring of employee salary protocols. The satellite hospital — which the agent's analysis had shown to be cash-positive if receivables were liquidated — was spun off and placed under an interim operational manager. The main hospital and the Howrah unit remained in resolution mode, with a 120-day timeline for a final resolution plan.

Shomit's chambers continued to run. He had three other cases in concurrent NCLT proceedings. One was a 2024 textile-mill insolvency (a Ballarpur Industries subsidiary in Bihar, falling under the Kolkata Bench's jurisdiction); another was a company scheme of arrangement for a Kolkata pharmaceutical firm. Each involved similar problems: precedent across years, creditor coordination across states, statutory compliance across multiple regulatory regimes.

What we hoped would happen is that Shomit — and advocates like him, in practices across India — would begin to think of a tool like this not as a research shortcut but as a structural necessity. A solo or small-firm advocate cannot afford a legal research team. The cost of legal research — the time to read case law, the overhead to maintain current precedent — is paid by a reduction in case volume or an increase in billable hours. The agent does not replace the advocate's judgment. It removes the cost of information-gathering, allowing the advocate to apply judgment to a larger caseload or to apply judgment more carefully to the cases he has.

For Shomit, the outcome has been modest but genuine: more time in the High Court arguing cases, less time in chambers chasing precedent. For Priya, it has been more profound: she now submits drafted pleadings with confidence that the legal reasoning is not silently misaligned with recent precedent, because the agent has verified that alignment. The spreadsheet she assembled — a document she thought was merely administrative — became, through the agent's reading, a legal instrument in its own right.

The High Court of Calcutta, founded in 1862, handles one of the largest caseloads of any superior court in the world. Its bar — the advocates, solicitors, junior counsel, and the institution of law practice itself — has evolved continuously for 164 years. It will evolve again. If the pattern that Shomit and Priya have found — an advocate using an agent to cross-check precedent and coordinate information across jurisdictions — becomes common, the nature of legal practice itself will change. The question becomes not whether the advocate knows the case law (he does not need to memorize it) but whether he can ask the right question of the agent and whether he can judge the answer.

1. ⚖️

   15 Mar 2026 — NCLT Petitions Filed

   Three hospitals in the RG-Kar network file insolvency petitions simultaneously at NCLT Kolkata Bench. Collective debt: ₹11.2 crores. Employee arrears: ₹1.8 crores.

2. 📨

   2 May 2026 — Spreadsheet Analysis

   Priya's 147-row creditor spreadsheet is analyzed by the agent, surfacing three overlooked creditors, four compliance gaps (including the ₹47,000 KMC property-tax claim), and a cash-flow reconstruction showing differential insolvency across the three hospitals.

3. 🛑

   16 May 2026 — Precedent Check

   Shomit's draft argument on receivables liquidation is cross-checked against NCLT Kolkata Bench precedent. IRP v. State Bank of India (July 2023) reveals a conflict-of-interest disclosure requirement. Argument is revised.

4. 💸

   18 May 2026 — Creditor Committee Approval

   Revised motion (requesting CoC written consent + disclosing state conflict-of-interest) passes. Satellite hospital receivables liquidation authorized. Cash flow impact: ₹82 lakhs recovered for employee-salary clearance.

5. ₹

   May–June 2026 — Resolution Plan Structured

   Creditor committee approves partial state equity injection + employee salary restructuring + satellite hospital spinoff. Main hospital and Howrah unit enter 120-day resolution-plan timeline.

"এক সময় আইনজীবীর মেধা ছিল তার মেমোরি — যে কত কেস তিনি পড়েছেন, কত নিয়ম মনে রেখেছেন। এখন এটা তার ফোকাসের ব্যাপার। কোন প্রশ্নটা অর্থপূর্ণ, কোন উত্তরটা সঠিক, কোন যুক্তিটা জেতে যাবে।"

— At one time, an advocate's intelligence was his memory — how many cases he had read, how many rules he had memorized. Now it is a matter of focus. Which question is worth asking, which answer is right, which argument wins.

The RG-Kar hospital network is free to use the agent. We have tailored the research module for Indian insolvency law, High Court precedent, and the specific quirks of NCLT Kolkata Bench jurisprudence. The cost to Shomit is zero. The cost to us is the hosting of the research model and the continuous refresh of case-law integration. We will not advertise to Shomit or Priya. We will not sell their case details. We will cross-check their arguments against precedent, one pleading at a time, and we will be quiet.

The question, for the next six months, is whether Shomit will use the agent for the textile-mill insolvency and the pharmaceutical company scheme. If precedent-checking becomes a reflex — something a junior advocate does automatically before submitting a pleading — the nature of that junior advocate's work changes. She is no longer a research conductor managing case-law databases; she is a strategic thinker, a writer, an arguer. The agent carries the burden of currency; she carries the burden of judgment. That is the division we hope will hold.