Claude for Legal — AI for Indian in-house counsel & transactional lawyers

The Reality

Two gaps. Both are fixable in a weekend.

Most lawyers who've tried AI tools haven't got reliable results. Not because the tools are bad — because neither gap has been addressed.

The knowledge gap. Generic tools aren't configured for Indian practice.

AI tools are powerful — but out of the box, they have no knowledge of your practice context, your risk thresholds, your firm's standard positions, or the specific obligations under DPDP, SEBI, MCA, BCI norms, and the Sakshya Adhiniyam that govern your work.

Without India-specific configuration, every output needs manual verification. You end up doing the work twice — which defeats the efficiency case entirely.

India-specific configuration required

The execution gap. Curiosity without workflow design gives inconsistent results.

Most lawyers have tried ChatGPT or Copilot. The results were inconsistent — sometimes impressive, sometimes wrong in ways that were hard to catch. The problem isn't the model. It's that ad-hoc prompting isn't a workflow.

Without a repeatable system — practice profile, prompt architecture, verification layer — you can't trust the output professionally. This bootcamp closes that gap.

Repeatable workflow required

Who It's For

For lawyers who want to turn volume into velocity.

The high-leverage work — advising, structuring, closing — keeps getting crowded out by volume work. This bootcamp addresses the volume. Specifically.

In-House Counsel

Corporate Legal Teams

Your inbox is a contract review queue. DPDP, SEBI, RBI, and MCA compliance all run through you. You're drafting policies, reviewing vendor agreements, and monitoring regulatory changes — often alone or with a very small team.

DPDP deadlines stop being a scramble. Vendor contracts take 20 minutes instead of two days. New SEBI circulars land in your inbox before you open your browser. You get your Fridays back.

Transactional Lawyer

M&A, PE, Contracts & Banking

Due diligence is still manual. Contract review is still a highlighter and a precedent file. First drafts still start from an old template. The high-leverage work you're good at gets consumed by the volume work below it.

DD cycles that used to take weeks are compressed significantly. Contract review that consumed your mornings takes 20 minutes. First drafts start in your firm's style, not a generic template. The high-leverage work rises back to the top.

Not designed for bespoke litigators. If your practice is primarily courtroom litigation — hearing preparation, pleadings, oral argument strategy — this cohort is not the right fit. The workflow is fundamentally different. We're building a separate track for litigation support. Email us → and we'll flag it when it's ready.

Curriculum

Two days. Every module produces something you keep.

Built for beginner AI users. No coding or terminal experience required. Every module ends with a concrete artifact built on your own practice.

Day 1 Saturday — AI-assisted document work

AI document workspace AI in Microsoft Word

Getting Started: Tools, Data Safety & Professional Responsibility How to configure AI safely for legal work — what the tools can see, what they can’t, and your obligations under BCI norms and the Advocates Act.

Personal prompt template you'll reuse in every module today

Prompt Architecture for Legal Work

How to write instructions that AI follows precisely and reliably — the difference between a vague prompt and a legal workflow you can trust.

Contract Review & Summarisation Key clauses extracted and summarised in plain English, structured by risk area — ready to paste into your tracker or send to a stakeholder.

Clause extraction output + self-verified audit, ready to paste into your contract tracker

Your Negotiating Playbook & AI-Assisted Redlining Store your standard positions and fallbacks in one document. AI redlines contracts against them and drafts the issues list.

Locked playbook .md + redlined contract + issues memo

Risk Review Across a Contract Portfolio Run the same analysis across 10, 20, or 50 contracts at once. Deviations from your standards, outliers, counterparty patterns — one report.

Risk register with deviations, outliers, and counterparty patterns

Where AI Makes Mistakes — and How You Catch Them Six failure modes shown live: hallucinated citations, missed clauses, version errors, false confidence, privilege exposure under Section 132 Sakshya Adhiniyam, and non-delegable review obligations.

One-page reference card: the six failure modes to watch for in every AI workflow

Building Your First Automated Legal Workflow No coding required. You describe what you want in plain English — the AI builds and runs it. If you can brief a junior lawyer, you can do this.

Claude Code running + CLAUDE.md configured for your practice

Day 2 Sunday — Automated legal workflows

AI workflow automation

Your AI Practice Profile A document that tells your AI assistant who you are, what law you practice, and what your standards are. Built once — every workflow runs on it automatically.

Production-ready CLAUDE.md for your practice

One-Click Review with Automatic Error-Checking Turn your Day 1 workflows into single commands. Add automatic verification that catches errors before you see the output.

One-command extraction + automatic verification running on every contract

Review an Entire Contract Portfolio Simultaneously Run extraction, redlining, verification, and summarisation across a full folder in one operation — the way a team of lawyers would, but in minutes.

Full review pipeline running in parallel on a contract folder — no context ceiling

Automated Regulatory Monitoring SEBI, RBI, MCA, DPDP — new circulars summarised and in your inbox before you open your browser, with source citations on every item.

Regulatory digest skill: what changed, who's affected, what to do — with source citations

AI Legal Q&A: Answers from Your Documents Only Ask any question. Get a cited answer drawn only from your approved policies or firm precedents — not the internet, not a guess.

AI advisory engine that answers from approved sources only — and escalates when unsure

Your Monday Plan One workflow. One document. One Monday it runs on. Day-7 and day-30 check-ins scheduled before you leave.

Monday plan locked into your CLAUDE.md + day-7 and day-30 check-ins in your calendar

The Shift

Monday morning, you lead differently.

Not productivity theatre. Structural changes to where your time actually goes — built in two days, running on Monday.

Before

4 hours reviewing a 40-page contract with a highlighter and a precedent file

Manually checking SEBI, MCA, and RBI portals every morning for new circulars

Ctrl+F through 200-page due diligence documents, building issue lists by hand

Adapting an old NDA template and hoping it's still current

Calendar reminder for DPDP deadline, followed by a last-minute scramble

After your AI stack

Red-flagged issue list in 20 minutes, benchmarked against your risk thresholds

New circulars surfaced to your inbox overnight, pre-summarised with applicability flags

Extracted issues cross-referenced to your DD checklist, with page citations

AI-drafted first cut in your firm's style, ready for partner review in 30 minutes

Live gap tracker with automated deadline alerts and remediation status

For in-house counsel: an AI workflow that saves 30 hours a month costs less to build than one month of a junior hire. For firm lawyers: recovering one additional deal cycle per quarter covers this bootcamp in the first month.

Instructors

Designed by a lawyer who lived these workflows.

Parivesh spent years at AZB & Partners and Khaitan & Co doing the exact work this bootcamp addresses. He designed every module around the specific tasks that consume partner and senior associate time in Indian transactional practice.

Parivesh Agrawal

Ex-AZB & Partners · Ex-Khaitan & Co.

Years at AZB & Partners and Khaitan & Co. doing M&A, PE, and corporate work. Designed the curriculum from the ground up, starting with the workflows that consume the most time in Indian legal practice. ISB · Symbiosis Law School.

Shameek Chakravarty

Co-founder, theCRUX

4X founder. 20+ years building products across startups and big tech — Amazon, Yahoo, Sun Microsystems, Juniper Networks. Built two AI ventures — Mantid.ai and Juvant.ai. Has trained 600+ professionals. CS BITS Pilani · MBA ISB.

Bharanidharan Viswanathan

Co-founder, theCRUX

3X founder. Built 91mobiles, SaaSworthy, and Intoglo across 20+ years in B2C, B2B, and cross-border logistics. Went deep on AI in 2021 — the models, the maths, and what separates real workflows from hype. Teaches with a builder's impatience for theory. CEG Anna University · MBA ISB.

FAQ

Common questions.

This cohort is designed for in-house counsel and transactional lawyers — M&A, PE, contracts, and compliance-heavy practice. Bespoke litigators have a fundamentally different workflow (hearing prep, pleadings, oral arguments), and the systems built here won't map cleanly to it. We're building a separate track for litigation support. Email hello@thecrux.ai and indicate your practice area — we'll flag the right programme when it opens.

No. Day 1 (Cowork + Claude for Word) requires zero technical background. Day 2 introduces Claude Code — but you'll be giving instructions in plain English, not writing Python. The tool takes your description and builds the software. If you can precisely describe what you want a junior lawyer to do, you can direct Claude Code. Precision of instruction is a legal skill. That's what matters here.

We cover data handling in detail on Day 1, before you build anything. The tools used in this bootcamp operate under Anthropic's data usage policies, and we walk through exactly what those mean for legal work — what gets processed, what doesn't, and how to configure each tool correctly for your practice context.

We also cover anonymisation protocols: removing client names, changing identifying details, keeping the legal substance. This is a 30-second habit that protects your clients regardless of platform settings. We set up your configuration correctly before you build anything substantive.

Yes, with the right guardrails — which we build into every workflow on Day 1. The BCI norms that apply to AI in legal practice are covered specifically, including the implications of Section 132 of the Sakshya Adhiniyam for AI-generated outputs used in proceedings. Every workflow we build has a lawyer-in-the-loop review layer. No AI output goes to a client without your judgement on it. That's not just an ethics requirement — it's also better practice.

The bootcamp is built on Claude (Anthropic), accessed via Cowork on Day 1, Claude for Word for document drafting, and Claude Code on Day 2. Anthropic's data processing terms are compatible with standard firm data handling policies — we review this on Day 1. You'll need a Claude Pro subscription (approximately ₹1,700/month, billed directly by Anthropic) and a laptop. Full pre-work checklist is sent with your confirmation.

Live, online, IST. Saturday and Sunday, 10 AM – 5 PM IST each day. The first cohort runs June 20–21, 2026. Under 30 seats — the small cohort is deliberate; it means instructors can work through your specific practice context, not just demo generic workflows.

The first cohort runs June 20–21, 2026 (Saturday & Sunday), 10 AM – 5 PM IST, live online. Seats are limited to under 30 — reserve early to lock in the early bird rate.

₹14,999 early bird while seats remain — regular price is ₹25,000. We issue a GST invoice; most participants expense this as L&D or professional development. Group pricing for 3+ seats — email hello@thecrux.ai. You'll also need a Claude Pro subscription (approximately ₹1,700/month, billed separately by Anthropic).

Yes. Group pricing is available for 3 or more seats. Email hello@thecrux.ai with the number of participants and your firm name — we'll set it up with a single PO and invoice. The workflows you build are entirely yours: your CLAUDE.md, your skills, your systems. No ongoing license required to use them in your practice after the bootcamp.

Yes. We issue a GST invoice. Most participants expense this as professional development or training. If your firm needs a purchase order process, email hello@thecrux.ai and we'll handle it directly. Group pricing is available for 3+ seats — see above.

Standard practice is to bill at your normal rate with a notation in the work description: "drafted with AI assistance, reviewed and verified by [name]." Some firms are experimenting with a modest efficiency discount on fixed-fee matters. This is covered explicitly in the Day 2 Wrap session — both the in-house angle (cost avoidance, not billing) and the firm angle (margin vs. rate implications). Check your engagement letter and discuss with your client contact before the first matter.

Sessions are recorded and available for lifetime reference. However, the core of what you build — systems trained on your own practice data, built through hands-on sessions with instructors in the room — can't be replicated from a recording. The cohort dynamic is a significant part of the product. Show up live.

Day-7 and Day-30 check-in sessions are scheduled in-room before you leave — calendar invites go out before anyone walks out the door. You have lifetime access to session recordings. If something you built breaks in Week 3 or you're stuck on a workflow, email hello@thecrux.ai — we'll help debug it. The goal is working systems on Monday, not a certificate on LinkedIn.

Become the AI-fluent lawyer in the room. In one weekend.

Two gaps. Both are fixable in a weekend.

The knowledge gap. Generic tools aren't configured for Indian practice.

The execution gap. Curiosity without workflow design gives inconsistent results.