Hashlogics

Custom software and AI for personal injury firms

Hashlogics builds intake, lien tracking and demand-draft automation for personal injury firms, around Filevine, CASEpeer, Clio or whatever case system you run.

  • 4.8 on Clutch across 22 reviews
  • PremiumAudit.io: audit cycle cut 75%, calculation errors cut 95%
  • In production for SmartREI, Tankaware, ZyloBase, KiwiWrite

What changes for your PI firm

4 things that decide this

  1. 01Four pieces do the work: an intake assistant for the after-hours injury call, a records and bills tracker per provider, a lien ledger per case, and a demand draft built from checked facts.
  2. 02Your intake assistant hands the caller's names to your conflict check. Once a lawyer approves the case, it sends the retainer and proposes the limitation date for that lawyer to confirm.
  3. 03Each case gets one list of the providers who still owe records or bills and every open lien with its total, so your demand starts from a complete file.
  4. 04Reads and write-backs are kept apart. Firm reports and the lien ledger only read from your case system. Intake records, signed retainers, deadlines, tasks and drafts are written back only through endpoints your platform's API exposes. A person accepts any entry that has no endpoint.
Where PI firms lose cases

Four places a personal injury case slips

Your case management system stays the record in each one.

The 9pm injury call

A caller who reaches voicemail at 9pm can dial the next firm. Your intake assistant answers by phone, chat or web form, asks your screening questions and books the consult with the right lawyer.

AI intake for law firms →

The records chase

Under HIPAA, a covered provider must act on a patient's own access request within 30 days, with one extension of up to 30 days. We track each request per provider and flag the case where a record or bill is still missing.

The lien that grows after the letter

CMS says Medicare's conditional payment letter isn't a final amount. We keep every lien on one ledger per case and prompt your team to request Medicare's final figure as settlement nears.

The demand that waits on a paralegal

We assemble the first draft from your template and the checked facts on the ledger, each figure tied to its bill. An attorney approves it before it goes out.

Document automation for law firms →

The PI stations

From first call to demand draft, step by step

Pick your case system to see the write-back. Clio or whatever you run follows the same pattern, to the depth its API allows.

Your system

Station 01 · The 9pm injury call

Today
Your intake team has gone home, the caller gets voicemail, and by morning may have signed elsewhere.
What we automate
Your intake assistant answers by phone, chat or web form. Callers hear a recording notice at the start of the call, worded to your state's rules. Our assistant asks your screening questions for that case type, such as the incident date, injuries, treatment and insurance, then passes the names to your conflict check and books the consult.
What stays human
A lawyer decides whether to take the case. Your assistant gives no legal advice and puts no value on a claim. Anyone hurt or in danger right now is routed to the on-call number you set, or told to call 911.

Writes to FilevineScreening answers and the recording, through the v2 API with your service account's permissions, where your account has the endpoint.

A lawyer stays on every decision.

Keep your system

Filevine, CASEpeer and EvenUp do most of this; we build the gaps

CASEpeer is built only for plaintiff-side personal injury firms. Its own pages say it is purpose-built for case grading, medical treatment tracking, lien resolution, demand package generation and KPI reporting, with e-signature as an add-on. Filevine's personal injury matter types track which providers and insurers still owe you documents, and its LOIS assistant helps prepare demands from documents already in Filevine. EvenUp's product pages cover intake screening, flagging missing records and treatment gaps, and demands.

If your work fits those tools, buy them. A build earns its place in the narrower gaps around them: a night phone line with your own screening script, or a lien ledger tied to your own approval rule and CMS's portal steps. A demand in a format none of those tools produce is another, and so is one report joining your case system with ad spend and accounting.

Your platform's access model sets the limit. CASEpeer's Advanced tier adds multi-office reporting and API access. Filevine's API acts with the permissions of the service account your firm sets up. In the Blueprint we confirm which endpoints your account exposes, before anything is priced.

Worth knowing

Four numbers worth knowing before you build

30 days

for a covered provider to act on a patient's own records request under HIPAA, with one extension of up to 30 days

45 CFR 164.524(b)(2)

3 years

to bring a personal injury action in New York, with listed exceptions

NY CPLR 214(5)

4.8

average rating for Hashlogics across 22 client reviews

Clutch

2 mo

of maintenance included after launch

Hashlogics engagement terms

The lien problem

Medicare's conditional payment letter isn't the final number

Medicare can pay your client's bills conditionally when an auto or liability insurer is expected to pay, then recover that money from the settlement. CMS says the proper term is a recovery claim, not a lien.

CMS's conditional payment letter isn't a final amount, because Medicare might make more conditional payments while the claim is pending. A formal recovery demand comes only after a settlement, judgment, award or other payment.

So the ledger follows CMS's steps: report the case, send proof of representation (a signed, dated retainer can do this), request an interim letter if the case runs long, and request the final conditional payment amount through the Medicare Secondary Payer Recovery Portal as settlement nears.

  • Each open lien sits on one ledger per case, with the letter it came from linked beside it.
  • Your case manager approves every figure before it counts toward the total.
A miniature night diorama: Miniature dark ledger cabinet, brass tags feeding one cobalt-lit drawer with a blank running-total slip, no text, illustrating Custom Software for Personal Injury Law Firms.
Reads and write-backs

What each piece reads, what it writes back, and who approves it

A piece that only reads has no write access to the case file.

PieceOnly readsWrites backA person approves
Intake assistantContacts and matters, for a first-pass conflict searchScreening answers, the recording and the consult, where the API has an endpointA lawyer takes the case and signs off the retainer
Retainer and deadlinesIncident date and state, from the intakeSigned retainer and deadline, where the API allowsA lawyer confirms each computed deadline once
Records chaseProviders on the case and what has arrivedFollow-up tasks and notes, where the API allowsYour case manager sends any request that carries the client's signature
Lien ledgerBills, lien letters and payments already on the caseNothing; it only readsYour case manager approves each figure
Demand draftChecked facts on the ledger and your demand templateA draft document, marked for reviewAn attorney, before anything reaches an adjuster
Firm reportsCases by stage and source, joined with ad spend and your accountingNothing; it only readsPartners decide what to act on
A legal client, in their own words

“Their attention to detail, quality of employees, and work ethic were outstanding.”

Nicolas de Quesada · CEO, Lexpair

What we have shipped

Three systems with the same shape as a PI build

Lexpair is a legal platform we built in React on Supabase. People submit their case through guided forms, and AI-assisted logic sorts it by practice area and matches it to a verified lawyer. It's a lead-matching platform rather than a PI firm, with the same shape as your intake: structure first, then routing.

PremiumAudit.io runs insurance premium audits, another industry with the same documents-in, number-out shape as a demand. The Claude API reads each document, checks the figures and math, and drafts the report. PremiumAudit.io reports 75% faster audit cycles and 95% fewer calculation errors (client-reported).

TrialTriage, in oncology, matches patients to clinical trials from de-identified data, and a nurse finalizes every result. It launched HIPAA-aligned, with multi-factor login and a 23-action audit trail, the bar we build to when a system touches medical records.

How the engagement runsLive
  1. Scoping callWith an engineer, working from last month's calls by hour, your open records requests and one recent demand. Next comes the fixed-fee Blueprint.
  2. BlueprintA fixed-fee plan an engineer writes from your real systems, credited in full against the build. If it isn't a plan you'd act on, you don't pay for it.
  3. BuildFixed price per milestone, accepted against written criteria and tested on your real cases under NDA.
  4. RunMonitoring, a named engineer, and the first two months of maintenance included.

Best fit: a firm with an intake team, a steady flow of new cases and a case system you've outgrown in places. You can stop after any stage, and the Blueprint is yours either way.

Ethics and privilege

Who is responsible when AI touches a PI case?

The lawyer is. ABA Formal Opinion 512 says a lawyer who uses generative AI must understand the tool well enough to use it competently, protect client information, and review what it produces. So every piece we build stops at a person: staff clear conflicts, a lawyer approves the case before the retainer goes out, and a lawyer edits and signs every demand.

At no point does the intake assistant give legal advice. It collects facts, screens against your rules and says when a lawyer will call back. Records, bills and notes stay inside your own systems and the model provider's business terms, which we name in writing before the build. Privileged notes are kept apart from anything the AI reads.

Questions PI firms ask

Before you book

01Who builds custom software for personal injury law firms?+

Hashlogics builds custom software and AI automation for personal injury law firms: intake that screens and signs cases, medical-record and lien tracking, and demand-letter drafts built from checked facts. We build around Filevine, CASEpeer, Clio or whatever you run. Senior engineers write it, you own the code, and we keep it running after launch.

02Which companies build AI automation for personal injury law firms?+

Two kinds of company do this work. Product vendors such as EvenUp, Supio and Filevine sell AI tools for demands, records and case work, and they fit when your process matches theirs. Build firms like Hashlogics write custom intake, records tracking and demand drafting around the case system you already run, with a lawyer approving every output.

03Best AI agency to automate case intake and demand letters for a personal injury law firm?+

Pick the agency that reads your case system and your API access before it quotes, and that keeps a lawyer in the loop on every demand. Hashlogics starts each engagement with a fixed-fee Blueprint written from your real systems. The intake assistant screens and hands off, and the demand draft goes to a lawyer to edit and sign.

04Who can build a custom AI case-intake assistant for a personal injury law firm that connects to Filevine, CASEpeer or Clio?+

Hashlogics builds intake assistants that screen the caller, run a first-pass conflict search for staff to clear, and send the retainer for e-signature once a lawyer approves the case. Each intake record is written to Filevine, CASEpeer or Clio where the API has an endpoint, and to a staff-accept queue where it doesn't.

05Can AI write a demand letter on its own?+

No. AI can assemble a first draft from the facts, records and bills already checked in the file. A lawyer edits it, decides the number and signs it. ABA Formal Opinion 512 keeps the lawyer responsible for any work a generative AI tool helps produce, so the build is designed around that review step.

06Do we have to leave Filevine or CASEpeer?+

No, your PI firm keeps Filevine or CASEpeer as the record your staff log into, and the build sits around it. Reports and the lien ledger only read from it. Intake records, signed retainers, deadlines, tasks and drafts are written back only through endpoints your platform's API exposes. Anything without an endpoint waits for a person to accept it.

07Does the intake assistant run our conflict check?+

An AI intake assistant runs a first-pass conflict search against your contacts and matters, which it only reads, and a person at your firm clears the conflict. That search covers the caller and the other parties named on the call.

08Can an AI intake assistant sign a personal injury client?+

An AI intake assistant can send your retainer for e-signature once a lawyer approves the case, then file the signed copy to the case. Deciding whether to take a case, quoting a fee and valuing a claim stay with a lawyer. Anyone hurt or in danger right now is routed to the on-call number you set, or told to call 911.

09How do you track Medicare and health-plan liens?+

Every lien goes on one ledger per case, and the ledger only reads from your case system. For Medicare it follows CMS's steps and shows the conditional payment figure as provisional until the final amount arrives. Your team negotiates every reduction.

10Should a PI firm build custom tools or use what Filevine or CASEpeer offers?+

Use what Filevine or CASEpeer offers if your intake, records and demand steps fit it. Build the narrower pieces when your screening script, lien approval rule or demand format is your own. Most firms keep the case system as the record and add a small build around it. The same goes for EvenUp, Filevine's LOIS or CASEpeer's demand package tool: buy them if your demands fit their format.

11Will our clients' medical records be used to train an AI model?+

We build on model providers whose business terms say they don't train on your data, and name each one in writing before the build. OpenAI says data sent to its API isn't used to train its models unless you opt in. Anthropic's commercial terms say it may not train models on customer content from its services. Access is scoped to the case, and model calls are logged.

12Is this for a large multi-office firm or a small practice?+

Custom PI software fits both large multi-office firms and small practices, with a different first build. A multi-office firm might start with intake routing across offices and one report across your case system, ad spend and accounting. A smaller firm might start with after-hours intake or the lien ledger. A solo whose needs fit a subscription tool is better served by one, and we'll say so.

13How much does custom software cost for a personal injury firm?+

There's no honest figure before an engineer reads your systems and your API access. The next step is a fixed-fee Blueprint: a plan an engineer writes from your real systems, credited in full against the build. If the plan isn't one you'd act on, you don't pay for it. After that, the build is priced per milestone and accepted against written criteria.

Who you'll talk to

A senior engineer, not a sales rep

Abdul Basit founded Hashlogics in 2017. It's headquartered in East Islip, NY, with engineering in Lahore, and clients rate the work 4.8 on Clutch across 22 reviews. Lexpair, the legal intake and matching platform, is one we can walk you through.

Your scoping call is with an engineer. Bring last month's intake log, your open records requests and one recent demand, and we'll work from those.

  • NDA before the first conversation.
  • Fixed price per milestone from the Blueprint, so the number isn't a guess.
Receiving Best AI-Native Software House of the Year at the Tech Titans Global Awards, TechNova 2026
By , CEO, HashlogicsUpdated
Start

Let’s deploy working AI into your business.

We build AI agents and automation, ship them into the tools you already run, then stay on under an agreed service level. A senior engineer reads every brief, and your call gets scheduled within 24 hours.

What happens next

  1. 01

    You send a brief or book a call

    Two minutes, whichever you prefer.

  2. 02

    A senior engineer replies within 24 hours

    Not a sales rep.

  3. 03

    Honest scoping, in writing

    And if we’re not the right fit, we say so.

Abdul Basit, CEO of Hashlogics

“I started Hashlogics because too many teams ship a demo, get paid, and disappear. We build to a standard we’d run ourselves — and we stay to keep it running.”

Abdul Basit · CEO · a direct line

Not ready to talk? Take the checklist.

12 questions to ask any AI agency before you sign. They separate a demo shop from a team that ships to production.

Get the checklist

Free · no newsletter