diff --git a/flake.lock b/flake.lock
index 897785ed0..1979a3d97 100644
--- a/flake.lock
+++ b/flake.lock
@@ -1,5 +1,26 @@
 {
   "nodes": {
+    "flake-parts": {
+      "inputs": {
+        "nixpkgs-lib": [
+          "nix-eval-jobs",
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1722555600,
+        "narHash": "sha256-XOQkdLafnb/p9ij77byFQjDf5m5QYl9b2REiVClC+x4=",
+        "owner": "hercules-ci",
+        "repo": "flake-parts",
+        "rev": "8471fe90ad337a8074e957b69ca4d0089218391d",
+        "type": "github"
+      },
+      "original": {
+        "owner": "hercules-ci",
+        "repo": "flake-parts",
+        "type": "github"
+      }
+    },
     "libgit2": {
       "flake": false,
       "locked": {
@@ -46,6 +67,30 @@
         "type": "github"
       }
     },
+    "nix-eval-jobs": {
+      "inputs": {
+        "flake-parts": "flake-parts",
+        "nix-github-actions": [],
+        "nixpkgs": [
+          "nixpkgs"
+        ],
+        "treefmt-nix": "treefmt-nix"
+      },
+      "locked": {
+        "lastModified": 1732550783,
+        "narHash": "sha256-HvIOV9Gf82Sc+WaQy7Wi9h8d5LXDkGwhbRUOmvGfEZM=",
+        "owner": "Ericson2314",
+        "repo": "nix-eval-jobs",
+        "rev": "bd5e93d30f87c936199af31d2448eb57b2357d57",
+        "type": "github"
+      },
+      "original": {
+        "owner": "Ericson2314",
+        "ref": "aggregate-jobs",
+        "repo": "nix-eval-jobs",
+        "type": "github"
+      }
+    },
     "nixpkgs": {
       "locked": {
         "lastModified": 1726688310,
@@ -66,8 +111,30 @@
       "inputs": {
         "libgit2": "libgit2",
         "nix": "nix",
+        "nix-eval-jobs": "nix-eval-jobs",
         "nixpkgs": "nixpkgs"
       }
+    },
+    "treefmt-nix": {
+      "inputs": {
+        "nixpkgs": [
+          "nix-eval-jobs",
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1723303070,
+        "narHash": "sha256-krGNVA30yptyRonohQ+i9cnK+CfCpedg6z3qzqVJcTs=",
+        "owner": "numtide",
+        "repo": "treefmt-nix",
+        "rev": "14c092e0326de759e16b37535161b3cb9770cea3",
+        "type": "github"
+      },
+      "original": {
+        "owner": "numtide",
+        "repo": "treefmt-nix",
+        "type": "github"
+      }
     }
   },
   "root": "root",
diff --git a/flake.nix b/flake.nix
index fccd45b96..048800c5b 100644
--- a/flake.nix
+++ b/flake.nix
@@ -8,6 +8,9 @@
   inputs.nix.inputs.nixpkgs.follows = "nixpkgs";
   inputs.nix.inputs.libgit2.follows = "libgit2";
 
+  inputs.nix-eval-jobs.url = "github:Ericson2314/nix-eval-jobs/aggregate-jobs";
+  inputs.nix-eval-jobs.inputs.nixpkgs.follows = "nixpkgs";
+
   # hide nix dev tooling from our lock file
   inputs.nix.inputs.flake-parts.follows = "";
   inputs.nix.inputs.git-hooks-nix.follows = "";
@@ -15,7 +18,10 @@
   inputs.nix.inputs.nixpkgs-23-11.follows = "";
   inputs.nix.inputs.flake-compat.follows = "";
 
-  outputs = { self, nixpkgs, nix, ... }:
+  # hide nix-eval-jobs dev tooling from our lock file
+  inputs.nix-eval-jobs.inputs.nix-github-actions.follows = "";
+
+  outputs = { self, nixpkgs, nix, nix-eval-jobs, ... }:
     let
       systems = [ "x86_64-linux" "aarch64-linux" ];
       forEachSystem = nixpkgs.lib.genAttrs systems;
@@ -26,6 +32,7 @@
       overlays.default = final: prev: {
         hydra = final.callPackage ./package.nix {
           inherit (nixpkgs.lib) fileset;
+          nix-eval-jobs = nix-eval-jobs.packages.${final.system}.default;
           rawSrc = self;
           nix-perl-bindings = final.nixComponents.nix-perl-bindings;
         };
@@ -69,6 +76,7 @@
       packages = forEachSystem (system: {
         hydra = nixpkgs.legacyPackages.${system}.callPackage ./package.nix {
           inherit (nixpkgs.lib) fileset;
+          nix-eval-jobs = nix-eval-jobs.packages.${system}.default;
           rawSrc = self;
           nix = nix.packages.${system}.nix;
           nix-perl-bindings = nix.hydraJobs.perlBindings.${system};
diff --git a/package.nix b/package.nix
index f944fe2b6..a9ec12c85 100644
--- a/package.nix
+++ b/package.nix
@@ -50,6 +50,7 @@
 , xz
 , gnutar
 , gnused
+, nix-eval-jobs
 
 , rpm
 , dpkg
@@ -190,6 +191,7 @@ stdenv.mkDerivation (finalAttrs: {
     openldap
     postgresql_13
     pixz
+    nix-eval-jobs
   ];
 
   checkInputs = [
@@ -218,6 +220,7 @@ stdenv.mkDerivation (finalAttrs: {
       darcs
       gnused
       breezy
+      nix-eval-jobs
     ] ++ lib.optionals stdenv.isLinux [ rpm dpkg cdrkit ]
   );
 
@@ -232,7 +235,7 @@ stdenv.mkDerivation (finalAttrs: {
   shellHook = ''
     pushd $(git rev-parse --show-toplevel) >/dev/null
 
-    PATH=$(pwd)/src/hydra-evaluator:$(pwd)/src/script:$(pwd)/src/hydra-eval-jobs:$(pwd)/src/hydra-queue-runner:$PATH
+    PATH=$(pwd)/src/hydra-evaluator:$(pwd)/src/script:$(pwd)/src/hydra-queue-runner:$PATH
     PERL5LIB=$(pwd)/src/lib:$PERL5LIB
     export HYDRA_HOME="$(pwd)/src/"
     mkdir -p .hydra-data
diff --git a/src/hydra-eval-jobs/hydra-eval-jobs.cc b/src/hydra-eval-jobs/hydra-eval-jobs.cc
deleted file mode 100644
index b83cae916..000000000
--- a/src/hydra-eval-jobs/hydra-eval-jobs.cc
+++ /dev/null
@@ -1,587 +0,0 @@
-#include <iostream>
-#include <thread>
-#include <optional>
-#include <unordered_map>
-
-#include "shared.hh"
-#include "store-api.hh"
-#include "eval.hh"
-#include "eval-gc.hh"
-#include "eval-inline.hh"
-#include "eval-settings.hh"
-#include "signals.hh"
-#include "terminal.hh"
-#include "util.hh"
-#include "get-drvs.hh"
-#include "globals.hh"
-#include "common-eval-args.hh"
-#include "flake/flakeref.hh"
-#include "flake/flake.hh"
-#include "attr-path.hh"
-#include "derivations.hh"
-#include "local-fs-store.hh"
-
-#include "hydra-config.hh"
-
-#include <sys/types.h>
-#include <sys/wait.h>
-#include <sys/resource.h>
-
-#include <nlohmann/json.hpp>
-
-void check_pid_status_nonblocking(pid_t check_pid)
-{
-    // Only check 'initialized' and known PID's
-    if (check_pid <= 0) { return; }
-
-    int wstatus = 0;
-    pid_t pid = waitpid(check_pid, &wstatus, WNOHANG);
-    // -1 = failure, WNOHANG: 0 = no change
-    if (pid <= 0) { return; }
-
-    std::cerr << "child process (" << pid << ") ";
-
-    if (WIFEXITED(wstatus)) {
-        std::cerr << "exited with status=" << WEXITSTATUS(wstatus) << std::endl;
-    } else if (WIFSIGNALED(wstatus)) {
-        std::cerr << "killed by signal=" << WTERMSIG(wstatus) << std::endl;
-    } else if (WIFSTOPPED(wstatus)) {
-        std::cerr << "stopped by signal=" << WSTOPSIG(wstatus) << std::endl;
-    } else if (WIFCONTINUED(wstatus)) {
-        std::cerr << "continued" << std::endl;
-    }
-}
-
-using namespace nix;
-
-static Path gcRootsDir;
-static size_t maxMemorySize;
-
-struct MyArgs : MixEvalArgs, MixCommonArgs, RootArgs
-{
-    Path releaseExpr;
-    bool flake = false;
-    bool dryRun = false;
-
-    MyArgs() : MixCommonArgs("hydra-eval-jobs")
-    {
-        addFlag({
-            .longName = "gc-roots-dir",
-            .description = "garbage collector roots directory",
-            .labels = {"path"},
-            .handler = {&gcRootsDir}
-        });
-
-        addFlag({
-            .longName = "dry-run",
-            .description = "don't create store derivations",
-            .handler = {&dryRun, true}
-        });
-
-        addFlag({
-            .longName = "flake",
-            .description = "build a flake",
-            .handler = {&flake, true}
-        });
-
-        expectArg("expr", &releaseExpr);
-    }
-};
-
-static MyArgs myArgs;
-
-static std::string queryMetaStrings(EvalState & state, PackageInfo & drv, const std::string & name, const std::string & subAttribute)
-{
-    Strings res;
-    std::function<void(Value & v)> rec;
-
-    rec = [&](Value & v) {
-        state.forceValue(v, noPos);
-        if (v.type() == nString)
-            res.emplace_back(v.string_view());
-        else if (v.isList())
-            for (unsigned int n = 0; n < v.listSize(); ++n)
-                rec(*v.listElems()[n]);
-        else if (v.type() == nAttrs) {
-            auto a = v.attrs()->find(state.symbols.create(subAttribute));
-            if (a != v.attrs()->end())
-                res.push_back(std::string(state.forceString(*a->value, a->pos, "while evaluating meta attributes")));
-        }
-    };
-
-    Value * v = drv.queryMeta(name);
-    if (v) rec(*v);
-
-    return concatStringsSep(", ", res);
-}
-
-static void worker(
-    EvalState & state,
-    Bindings & autoArgs,
-    AutoCloseFD & to,
-    AutoCloseFD & from)
-{
-    Value vTop;
-
-    if (myArgs.flake) {
-        using namespace flake;
-
-        auto [flakeRef, fragment, outputSpec] = parseFlakeRefWithFragmentAndExtendedOutputsSpec(fetchSettings, myArgs.releaseExpr, absPath("."));
-
-        auto vFlake = state.allocValue();
-
-        auto lockedFlake = lockFlake(
-                flakeSettings,
-                state,
-                flakeRef,
-            LockFlags {
-                .updateLockFile = false,
-                .useRegistries = false,
-                .allowUnlocked = false,
-            });
-
-        callFlake(state, lockedFlake, *vFlake);
-
-        auto vOutputs = vFlake->attrs()->get(state.symbols.create("outputs"))->value;
-        state.forceValue(*vOutputs, noPos);
-
-        auto aHydraJobs = vOutputs->attrs()->get(state.symbols.create("hydraJobs"));
-        if (!aHydraJobs)
-            aHydraJobs = vOutputs->attrs()->get(state.symbols.create("checks"));
-        if (!aHydraJobs)
-            throw Error("flake '%s' does not provide any Hydra jobs or checks", flakeRef);
-
-        vTop = *aHydraJobs->value;
-
-    } else {
-        state.evalFile(lookupFileArg(state, myArgs.releaseExpr), vTop);
-    }
-
-    auto vRoot = state.allocValue();
-    state.autoCallFunction(autoArgs, vTop, *vRoot);
-
-    while (true) {
-        /* Wait for the master to send us a job name. */
-        writeLine(to.get(), "next");
-
-        auto s = readLine(from.get());
-        if (s == "exit") break;
-        if (!hasPrefix(s, "do ")) abort();
-        std::string attrPath(s, 3);
-
-        debug("worker process %d at '%s'", getpid(), attrPath);
-
-        /* Evaluate it and send info back to the master. */
-        nlohmann::json reply;
-
-        try {
-            auto vTmp = findAlongAttrPath(state, attrPath, autoArgs, *vRoot).first;
-
-            auto v = state.allocValue();
-            state.autoCallFunction(autoArgs, *vTmp, *v);
-
-            if (auto drv = getDerivation(state, *v, false)) {
-
-                // CA derivations do not have static output paths, so we
-                // have to defensively not query output paths in case we
-                // encounter one.
-                PackageInfo::Outputs outputs = drv->queryOutputs(
-                    !experimentalFeatureSettings.isEnabled(Xp::CaDerivations));
-
-                if (drv->querySystem() == "unknown")
-                    state.error<EvalError>("derivation must have a 'system' attribute").debugThrow();
-
-                auto drvPath = state.store->printStorePath(drv->requireDrvPath());
-
-                nlohmann::json job;
-
-                job["nixName"] = drv->queryName();
-                job["system"] =drv->querySystem();
-                job["drvPath"] = drvPath;
-                job["description"] = drv->queryMetaString("description");
-                job["license"] = queryMetaStrings(state, *drv, "license", "shortName");
-                job["homepage"] = drv->queryMetaString("homepage");
-                job["maintainers"] = queryMetaStrings(state, *drv, "maintainers", "email");
-                job["schedulingPriority"] = drv->queryMetaInt("schedulingPriority", 100);
-                job["timeout"] = drv->queryMetaInt("timeout", 36000);
-                job["maxSilent"] = drv->queryMetaInt("maxSilent", 7200);
-                job["isChannel"] = drv->queryMetaBool("isHydraChannel", false);
-
-                /* If this is an aggregate, then get its constituents. */
-                auto a = v->attrs()->get(state.symbols.create("_hydraAggregate"));
-                if (a && state.forceBool(*a->value, a->pos, "while evaluating the `_hydraAggregate` attribute")) {
-                    auto a = v->attrs()->get(state.symbols.create("constituents"));
-                    if (!a)
-                        state.error<EvalError>("derivation must have a ‘constituents’ attribute").debugThrow();
-
-                    NixStringContext context;
-                    state.coerceToString(a->pos, *a->value, context, "while evaluating the `constituents` attribute", true, false);
-                    for (auto & c : context)
-                        std::visit(overloaded {
-                            [&](const NixStringContextElem::Built & b) {
-                                job["constituents"].push_back(b.drvPath->to_string(*state.store));
-                            },
-                            [&](const NixStringContextElem::Opaque & o) {
-                            },
-                            [&](const NixStringContextElem::DrvDeep & d) {
-                            },
-                        }, c.raw);
-
-                    state.forceList(*a->value, a->pos, "while evaluating the `constituents` attribute");
-                    for (unsigned int n = 0; n < a->value->listSize(); ++n) {
-                        auto v = a->value->listElems()[n];
-                        state.forceValue(*v, noPos);
-                        if (v->type() == nString)
-                            job["namedConstituents"].push_back(v->string_view());
-                    }
-                }
-
-                /* Register the derivation as a GC root.  !!! This
-                   registers roots for jobs that we may have already
-                   done. */
-                auto localStore = state.store.dynamic_pointer_cast<LocalFSStore>();
-                if (gcRootsDir != "" && localStore) {
-                    Path root = gcRootsDir + "/" + std::string(baseNameOf(drvPath));
-                    if (!pathExists(root))
-                        localStore->addPermRoot(localStore->parseStorePath(drvPath), root);
-                }
-
-                nlohmann::json out;
-                for (auto & [outputName, optOutputPath] : outputs) {
-                    if (optOutputPath) {
-                        out[outputName] = state.store->printStorePath(*optOutputPath);
-                    } else {
-                        // See the `queryOutputs` call above; we should
-                        // not encounter missing output paths otherwise.
-                        assert(experimentalFeatureSettings.isEnabled(Xp::CaDerivations));
-                        out[outputName] = nullptr;
-                    }
-                }
-                job["outputs"] = std::move(out);
-                reply["job"] = std::move(job);
-            }
-
-            else if (v->type() == nAttrs) {
-                auto attrs = nlohmann::json::array();
-                StringSet ss;
-                for (auto & i : v->attrs()->lexicographicOrder(state.symbols)) {
-                    std::string name(state.symbols[i->name]);
-                    if (name.find(' ') != std::string::npos) {
-                        printError("skipping job with illegal name '%s'", name);
-                        continue;
-                    }
-                    attrs.push_back(name);
-                }
-                reply["attrs"] = std::move(attrs);
-            }
-
-            else if (v->type() == nNull)
-                ;
-
-            else state.error<TypeError>("attribute '%s' is %s, which is not supported", attrPath, showType(*v)).debugThrow();
-
-        } catch (EvalError & e) {
-            auto msg = e.msg();
-            // Transmits the error we got from the previous evaluation
-            // in the JSON output.
-            reply["error"] = filterANSIEscapes(msg, true);
-            // Don't forget to print it into the STDERR log, this is
-            // what's shown in the Hydra UI.
-            printError(msg);
-        }
-
-        writeLine(to.get(), reply.dump());
-
-        /* If our RSS exceeds the maximum, exit. The master will
-           start a new process. */
-        struct rusage r;
-        getrusage(RUSAGE_SELF, &r);
-        if ((size_t) r.ru_maxrss > maxMemorySize * 1024) break;
-    }
-
-    writeLine(to.get(), "restart");
-}
-
-int main(int argc, char * * argv)
-{
-    /* Prevent undeclared dependencies in the evaluation via
-       $NIX_PATH. */
-    unsetenv("NIX_PATH");
-
-    return handleExceptions(argv[0], [&]() {
-
-        auto config = std::make_unique<HydraConfig>();
-
-        auto nrWorkers = config->getIntOption("evaluator_workers", 1);
-        maxMemorySize = config->getIntOption("evaluator_max_memory_size", 4096);
-
-        initNix();
-        initGC();
-
-        myArgs.parseCmdline(argvToStrings(argc, argv));
-
-        auto pureEval = config->getBoolOption("evaluator_pure_eval", myArgs.flake);
-
-        /* FIXME: The build hook in conjunction with import-from-derivation is causing "unexpected EOF" during eval */
-        settings.builders = "";
-
-        /* Prevent access to paths outside of the Nix search path and
-           to the environment. */
-        evalSettings.restrictEval = true;
-
-        /* When building a flake, use pure evaluation (no access to
-           'getEnv', 'currentSystem' etc. */
-        evalSettings.pureEval = pureEval;
-
-        if (myArgs.dryRun) settings.readOnlyMode = true;
-
-        if (myArgs.releaseExpr == "") throw UsageError("no expression specified");
-
-        if (gcRootsDir == "") printMsg(lvlError, "warning: `--gc-roots-dir' not specified");
-
-        struct State
-        {
-            std::set<std::string> todo{""};
-            std::set<std::string> active;
-            nlohmann::json jobs;
-            std::exception_ptr exc;
-        };
-
-        std::condition_variable wakeup;
-
-        Sync<State> state_;
-
-        /* Start a handler thread per worker process. */
-        auto handler = [&]()
-        {
-            pid_t pid = -1;
-            try {
-                AutoCloseFD from, to;
-
-                while (true) {
-
-                    /* Start a new worker process if necessary. */
-                    if (pid == -1) {
-                        Pipe toPipe, fromPipe;
-                        toPipe.create();
-                        fromPipe.create();
-                        pid = startProcess(
-                            [&,
-                             to{std::make_shared<AutoCloseFD>(std::move(fromPipe.writeSide))},
-                             from{std::make_shared<AutoCloseFD>(std::move(toPipe.readSide))}
-                            ]()
-                            {
-                                try {
-                                    auto evalStore = myArgs.evalStoreUrl
-                                             ? openStore(*myArgs.evalStoreUrl)
-                                             : openStore();
-                                    EvalState state(myArgs.lookupPath,
-                                            evalStore, fetchSettings, evalSettings);
-                                    Bindings & autoArgs = *myArgs.getAutoArgs(state);
-                                    worker(state, autoArgs, *to, *from);
-                                } catch (Error & e) {
-                                    nlohmann::json err;
-                                    auto msg = e.msg();
-                                    err["error"] = filterANSIEscapes(msg, true);
-                                    printError(msg);
-                                    writeLine(to->get(), err.dump());
-                                    // Don't forget to print it into the STDERR log, this is
-                                    // what's shown in the Hydra UI.
-                                    writeLine(to->get(), "restart");
-                                }
-                            },
-                            ProcessOptions { .allowVfork = false });
-                        from = std::move(fromPipe.readSide);
-                        to = std::move(toPipe.writeSide);
-                        debug("created worker process %d", pid);
-                    }
-
-                    /* Check whether the existing worker process is still there. */
-                    auto s = readLine(from.get());
-                    if (s == "restart") {
-                        pid = -1;
-                        continue;
-                    } else if (s != "next") {
-                        auto json = nlohmann::json::parse(s);
-                        throw Error("worker error: %s", (std::string) json["error"]);
-                    }
-
-                    /* Wait for a job name to become available. */
-                    std::string attrPath;
-
-                    while (true) {
-                        checkInterrupt();
-                        auto state(state_.lock());
-                        if ((state->todo.empty() && state->active.empty()) || state->exc) {
-                            writeLine(to.get(), "exit");
-                            return;
-                        }
-                        if (!state->todo.empty()) {
-                            attrPath = *state->todo.begin();
-                            state->todo.erase(state->todo.begin());
-                            state->active.insert(attrPath);
-                            break;
-                        } else
-                            state.wait(wakeup);
-                    }
-
-                    /* Tell the worker to evaluate it. */
-                    writeLine(to.get(), "do " + attrPath);
-
-                    /* Wait for the response. */
-                    auto response = nlohmann::json::parse(readLine(from.get()));
-
-                    /* Handle the response. */
-                    StringSet newAttrs;
-
-                    if (response.find("job") != response.end()) {
-                        auto state(state_.lock());
-                        state->jobs[attrPath] = response["job"];
-                    }
-
-                    if (response.find("attrs") != response.end()) {
-                        for (auto & i : response["attrs"]) {
-                            std::string path = i;
-                            if (path.find(".") != std::string::npos){
-                                path = "\"" + path  + "\"";
-                            }
-                            auto s = (attrPath.empty() ? "" : attrPath + ".") + (std::string) path;
-                            newAttrs.insert(s);
-                        }
-                    }
-
-                    if (response.find("error") != response.end()) {
-                        auto state(state_.lock());
-                        state->jobs[attrPath]["error"] = response["error"];
-                    }
-
-                    /* Add newly discovered job names to the queue. */
-                    {
-                        auto state(state_.lock());
-                        state->active.erase(attrPath);
-                        for (auto & s : newAttrs)
-                            state->todo.insert(s);
-                        wakeup.notify_all();
-                    }
-                }
-            } catch (...) {
-                check_pid_status_nonblocking(pid);
-                auto state(state_.lock());
-                state->exc = std::current_exception();
-                wakeup.notify_all();
-            }
-        };
-
-        std::vector<std::thread> threads;
-        for (size_t i = 0; i < nrWorkers; i++)
-            threads.emplace_back(std::thread(handler));
-
-        for (auto & thread : threads)
-            thread.join();
-
-        auto state(state_.lock());
-
-        if (state->exc)
-            std::rethrow_exception(state->exc);
-
-        /* For aggregate jobs that have named consistuents
-           (i.e. constituents that are a job name rather than a
-           derivation), look up the referenced job and add it to the
-           dependencies of the aggregate derivation. */
-        auto store = openStore();
-
-        for (auto i = state->jobs.begin(); i != state->jobs.end(); ++i) {
-            auto jobName = i.key();
-            auto & job = i.value();
-
-            auto named = job.find("namedConstituents");
-            if (named == job.end()) continue;
-
-            std::unordered_map<std::string, std::string> brokenJobs;
-            auto getNonBrokenJobOrRecordError = [&brokenJobs, &jobName, &state](
-                    const std::string & childJobName) -> std::optional<nlohmann::json> {
-                auto childJob = state->jobs.find(childJobName);
-                if (childJob == state->jobs.end()) {
-                    printError("aggregate job '%s' references non-existent job '%s'", jobName, childJobName);
-                    brokenJobs[childJobName] = "does not exist";
-                    return std::nullopt;
-                }
-                if (childJob->find("error") != childJob->end()) {
-                    std::string error = (*childJob)["error"];
-                    printError("aggregate job '%s' references broken job '%s': %s", jobName, childJobName, error);
-                    brokenJobs[childJobName] = error;
-                    return std::nullopt;
-                }
-                return *childJob;
-            };
-
-            if (myArgs.dryRun) {
-                for (std::string jobName2 : *named) {
-                    auto job2 = getNonBrokenJobOrRecordError(jobName2);
-                    if (!job2) {
-                        continue;
-                    }
-                    std::string drvPath2 = (*job2)["drvPath"];
-                    job["constituents"].push_back(drvPath2);
-                }
-            } else {
-                auto drvPath = store->parseStorePath((std::string) job["drvPath"]);
-                auto drv = store->readDerivation(drvPath);
-
-                for (std::string jobName2 : *named) {
-                    auto job2 = getNonBrokenJobOrRecordError(jobName2);
-                    if (!job2) {
-                        continue;
-                    }
-                    auto drvPath2 = store->parseStorePath((std::string) (*job2)["drvPath"]);
-                    auto drv2 = store->readDerivation(drvPath2);
-                    job["constituents"].push_back(store->printStorePath(drvPath2));
-                    drv.inputDrvs.map[drvPath2].value = {drv2.outputs.begin()->first};
-                }
-
-                if (brokenJobs.empty()) {
-                    std::string drvName(drvPath.name());
-                    assert(hasSuffix(drvName, drvExtension));
-                    drvName.resize(drvName.size() - drvExtension.size());
-
-                    auto hashModulo = hashDerivationModulo(*store, drv, true);
-                    if (hashModulo.kind != DrvHash::Kind::Regular) continue;
-                    auto h = hashModulo.hashes.find("out");
-                    if (h == hashModulo.hashes.end()) continue;
-                    auto outPath = store->makeOutputPath("out", h->second, drvName);
-                    drv.env["out"] = store->printStorePath(outPath);
-                    drv.outputs.insert_or_assign("out", DerivationOutput::InputAddressed { .path = outPath });
-                    auto newDrvPath = store->printStorePath(writeDerivation(*store, drv));
-
-                    debug("rewrote aggregate derivation %s -> %s", store->printStorePath(drvPath), newDrvPath);
-
-                    job["drvPath"] = newDrvPath;
-                    job["outputs"]["out"] = store->printStorePath(outPath);
-                }
-            }
-
-            job.erase("namedConstituents");
-
-            /* Register the derivation as a GC root.  !!! This
-                registers roots for jobs that we may have already
-                done. */
-            auto localStore = store.dynamic_pointer_cast<LocalFSStore>();
-            if (gcRootsDir != "" && localStore) {
-                auto drvPath = job["drvPath"].get<std::string>();
-                Path root = gcRootsDir + "/" + std::string(baseNameOf(drvPath));
-                if (!pathExists(root))
-                    localStore->addPermRoot(localStore->parseStorePath(drvPath), root);
-            }
-
-            if (!brokenJobs.empty()) {
-                std::stringstream ss;
-                for (const auto& [jobName, error] : brokenJobs) {
-                    ss << jobName << ": " << error << "\n";
-                }
-                job["error"] = ss.str();
-            }
-        }
-
-        std::cout << state->jobs.dump(2) << "\n";
-    });
-}
diff --git a/src/hydra-eval-jobs/meson.build b/src/hydra-eval-jobs/meson.build
deleted file mode 100644
index 916212e16..000000000
--- a/src/hydra-eval-jobs/meson.build
+++ /dev/null
@@ -1,8 +0,0 @@
-hydra_eval_jobs = executable('hydra-eval-jobs',
-  'hydra-eval-jobs.cc',
-  dependencies: [
-    libhydra_dep,
-    nix_dep,
-  ],
-  install: true,
-)
diff --git a/src/meson.build b/src/meson.build
index 8c7562ed2..52b821bc1 100644
--- a/src/meson.build
+++ b/src/meson.build
@@ -1,6 +1,5 @@
 # Native code
 subdir('libhydra')
-subdir('hydra-eval-jobs')
 subdir('hydra-evaluator')
 subdir('hydra-queue-runner')
 
diff --git a/src/script/hydra-eval-jobset b/src/script/hydra-eval-jobset
index 72a386f57..6dc1514b0 100755
--- a/src/script/hydra-eval-jobset
+++ b/src/script/hydra-eval-jobset
@@ -17,6 +17,7 @@ use Hydra::Helper::Nix;
 use Hydra::Model::DB;
 use Hydra::Plugin;
 use Hydra::Schema;
+use IPC::Run;
 use JSON::MaybeXS;
 use Net::Statsd;
 use Nix::Store;
@@ -357,22 +358,24 @@ sub evalJobs {
     my @cmd;
 
     if (defined $flakeRef) {
-        @cmd = ("hydra-eval-jobs",
-                "--flake", $flakeRef,
-                "--gc-roots-dir", getGCRootsDir,
-                "--max-jobs", 1);
+        @cmd = ("nix-eval-jobs", "--flake", $flakeRef . '#hydraJobs');
     } else {
         my $nixExprInput = $inputInfo->{$nixExprInputName}->[0]
             or die "cannot find the input containing the job expression\n";
 
-        @cmd = ("hydra-eval-jobs",
+        @cmd = ("nix-eval-jobs",
                 "<" . $nixExprInputName . "/" . $nixExprPath . ">",
-                "--gc-roots-dir", getGCRootsDir,
-                "--max-jobs", 1,
                 inputsToArgs($inputInfo));
     }
 
-    push @cmd, "--no-allow-import-from-derivation" if $config->{allow_import_from_derivation} // "true" ne "true";
+    push @cmd, ("--gc-roots-dir", getGCRootsDir);
+    push @cmd, ("--max-jobs", 1);
+    push @cmd, "--meta";
+    push @cmd, "--constituents";
+    push @cmd, "--force-recurse";
+    push @cmd, ("--option", "allow-import-from-derivation", "false") if $config->{allow_import_from_derivation} // "true" ne "true";
+    push @cmd, ("--workers", $config->{evaluator_workers} // 1);
+    push @cmd, ("--max-memory-size", $config->{evaluator_max_memory_size} // 4096);
 
     if (defined $ENV{'HYDRA_DEBUG'}) {
         sub escape {
@@ -384,14 +387,33 @@ sub evalJobs {
         print STDERR "evaluator: @escaped\n";
     }
 
-    (my $res, my $jobsJSON, my $stderr) = captureStdoutStderr(21600, @cmd);
-    die "hydra-eval-jobs returned " . ($res & 127 ? "signal $res" : "exit code " . ($res >> 8))
-        . ":\n" . ($stderr ? decode("utf-8", $stderr) : "(no output)\n")
-        if $res;
+    my $evalProc = IPC::Run::start \@cmd,
+            '>', IPC::Run::new_chunker, \my $out,
+            '2>', \my $err;
+
+    return sub {
+        while (1) {
+            $evalProc->pump;
+            if (!defined $out && !defined $err) {
+                $evalProc->finish;
+                if ($?) {
+                    die "nix-eval-jobs returned " . ($? & 127 ? "signal $?" : "exit code " . ($? >> 8)) . "\n";
+                }
+                return;
+            }
 
-    print STDERR "$stderr";
+            if (defined $err) {
+                print STDERR "$err";
+                undef $err;
+            }
 
-    return decode_json($jobsJSON);
+            if (defined $out && $out ne '') {
+                my $job = decode_json($out);
+                undef $out;
+                return $job;
+            }
+        }
+    };
 }
 
 
@@ -420,7 +442,7 @@ sub checkBuild {
     my $firstOutputName = $outputNames[0];
     my $firstOutputPath = $buildInfo->{outputs}->{$firstOutputName};
 
-    my $jobName = $buildInfo->{jobName} or die;
+    my $jobName = $buildInfo->{attr} or die;
     my $drvPath = $buildInfo->{drvPath} or die;
 
     my $build;
@@ -474,9 +496,30 @@ sub checkBuild {
 
         my $time = time();
 
-        sub null {
-            my ($s) = @_;
-            return $s eq "" ? undef : $s;
+        sub getMeta {
+            my ($s, $def) = @_;
+            return ($s || "") eq "" ? $def : $s;
+        }
+
+        sub getMetaStrings {
+            my ($v, $k, $acc) = @_;
+            my $t = ref $v;
+
+            if ($t eq 'HASH') {
+                push @$acc, $v->{$k} if exists $v->{$k};
+            } elsif ($t eq 'ARRAY') {
+                getMetaStrings($_, $k, $acc) foreach @$v;
+            } elsif (defined $v) {
+                push @$acc, $v;
+            }
+        }
+
+        sub getMetaConcatStrings {
+            my ($v, $k) = @_;
+
+            my @strings;
+            getMetaStrings($v, $k, \@strings);
+            return join(", ", @strings) || undef;
         }
 
         # Add the build to the database.
@@ -484,19 +527,19 @@ sub checkBuild {
             { timestamp => $time
             , jobset_id => $jobset->id
             , job => $jobName
-            , description => null($buildInfo->{description})
-            , license => null($buildInfo->{license})
-            , homepage => null($buildInfo->{homepage})
-            , maintainers => null($buildInfo->{maintainers})
-            , maxsilent => $buildInfo->{maxSilent}
-            , timeout => $buildInfo->{timeout}
-            , nixname => $buildInfo->{nixName}
+            , description => getMeta($buildInfo->{meta}->{description}, undef)
+            , license => getMetaConcatStrings($buildInfo->{meta}->{license}, "shortName")
+            , homepage => getMeta($buildInfo->{meta}->{homepage}, undef)
+            , maintainers => getMetaConcatStrings($buildInfo->{meta}->{maintainers}, "email")
+            , maxsilent => getMeta($buildInfo->{meta}->{maxSilent}, 7200)
+            , timeout => getMeta($buildInfo->{meta}->{timeout}, 36000)
+            , nixname => $buildInfo->{name}
             , drvpath => $drvPath
             , system => $buildInfo->{system}
-            , priority => $buildInfo->{schedulingPriority}
+            , priority => getMeta($buildInfo->{meta}->{schedulingPriority}, 100)
             , finished => 0
             , iscurrent => 1
-            , ischannel => $buildInfo->{isChannel}
+            , ischannel => getMeta($buildInfo->{meta}->{isChannel}, 0)
             });
 
         $build->buildoutputs->create({ name => $_, path => $buildInfo->{outputs}->{$_} })
@@ -665,7 +708,7 @@ sub checkJobsetWrapped {
         return;
     }
 
-    # Hash the arguments to hydra-eval-jobs and check the
+    # Hash the arguments to nix-eval-jobs and check the
     # JobsetInputHashes to see if the previous evaluation had the same
     # inputs.  If so, bail out.
     my @args = ($jobset->nixexprinput // "", $jobset->nixexprpath // "", inputsToArgs($inputInfo));
@@ -687,19 +730,12 @@ sub checkJobsetWrapped {
 
     # Evaluate the job expression.
     my $evalStart = clock_gettime(CLOCK_MONOTONIC);
-    my $jobs = evalJobs($project->name . ":" . $jobset->name, $inputInfo, $jobset->nixexprinput, $jobset->nixexprpath, $flakeRef);
-    my $evalStop = clock_gettime(CLOCK_MONOTONIC);
-
-    if ($jobsetsJobset) {
-        my @keys = keys %$jobs;
-        die "The .jobsets jobset must only have a single job named 'jobsets'"
-            unless (scalar @keys) == 1 && $keys[0] eq "jobsets";
-    }
-    Net::Statsd::timing("hydra.evaluator.eval_time", int(($evalStop - $evalStart) * 1000));
+    my $evalStop;
+    my $jobsIter = evalJobs($project->name . ":" . $jobset->name, $inputInfo, $jobset->nixexprinput, $jobset->nixexprpath, $flakeRef);
 
     if ($dryRun) {
-        foreach my $name (keys %{$jobs}) {
-            my $job = $jobs->{$name};
+        while (defined(my $job = $jobsIter->())) {
+            my $name = $job->{attr};
             if (defined $job->{drvPath}) {
                 print STDERR "good job $name: $job->{drvPath}\n";
             } else {
@@ -709,36 +745,20 @@ sub checkJobsetWrapped {
         return;
     }
 
-    die "Jobset contains a job with an empty name. Make sure the jobset evaluates to an attrset of jobs.\n"
-        if defined $jobs->{""};
-
-    $jobs->{$_}->{jobName} = $_ for keys %{$jobs};
-
-    my $jobOutPathMap = {};
-    my $jobsetChanged = 0;
-    my $dbStart = clock_gettime(CLOCK_MONOTONIC);
-
-
     # Store the error messages for jobs that failed to evaluate.
     my $evaluationErrorTime = time;
     my $evaluationErrorMsg = "";
-    foreach my $job (values %{$jobs}) {
-        next unless defined $job->{error};
-        $evaluationErrorMsg .=
-            ($job->{jobName} ne "" ? "in job ‘$job->{jobName}’" : "at top-level") .
-            ":\n" . $job->{error} . "\n\n";
-    }
-    setJobsetError($jobset, $evaluationErrorMsg, $evaluationErrorTime);
-
     my $evaluationErrorRecord = $db->resultset('EvaluationErrors')->create(
         { errormsg => $evaluationErrorMsg
         , errortime => $evaluationErrorTime
         }
     );
 
+    my $jobOutPathMap = {};
+    my $jobsetChanged = 0;
     my %buildMap;
-    $db->txn_do(sub {
 
+    $db->txn_do(sub {
         my $prevEval = getPrevJobsetEval($db, $jobset, 1);
 
         # Clear the "current" flag on all builds.  Since we're in a
@@ -751,7 +771,7 @@ sub checkJobsetWrapped {
             , evaluationerror => $evaluationErrorRecord
             , timestamp => time
             , checkouttime => abs(int($checkoutStop - $checkoutStart))
-            , evaltime => abs(int($evalStop - $evalStart))
+            , evaltime => 0
             , hasnewbuilds => 0
             , nrbuilds => 0
             , flake => $flakeRef
@@ -759,11 +779,24 @@ sub checkJobsetWrapped {
             , nixexprpath => $jobset->nixexprpath
             });
 
-        # Schedule each successfully evaluated job.
-        foreach my $job (permute(values %{$jobs})) {
-            next if defined $job->{error};
-            #print STDERR "considering job " . $project->name, ":", $jobset->name, ":", $job->{jobName} . "\n";
-            checkBuild($db, $jobset, $ev, $inputInfo, $job, \%buildMap, $prevEval, $jobOutPathMap, $plugins);
+        my @jobsWithConstituents;
+
+        while (defined(my $job = $jobsIter->())) {
+            if ($jobsetsJobset) {
+                die "The .jobsets jobset must only have a single job named 'jobsets'"
+                    unless $job->{attr} eq "jobsets";
+            }
+
+            $evaluationErrorMsg .=
+                ($job->{attr} ne "" ? "in job ‘$job->{attr}’" : "at top-level") .
+                ":\n" . $job->{error} . "\n\n" if defined $job->{error};
+
+            checkBuild($db, $jobset, $ev, $inputInfo, $job, \%buildMap, $prevEval, $jobOutPathMap, $plugins)
+                unless defined $job->{error};
+
+            if (defined $job->{constituents}) {
+                push @jobsWithConstituents, $job;
+            }
         }
 
         # Have any builds been added or removed since last time?
@@ -795,27 +828,26 @@ sub checkJobsetWrapped {
                 my $x = $buildMap{$id};
                 my $y = $drvPathToId{$x->{drvPath}};
                 if (defined $y) {
-                    next if length $x->{jobName} > length $y->{jobName};
-                    next if length $x->{jobName} == length $y->{jobName} && $x->{jobName} ge $y->{jobName};
+                    next if length $x->{attr} > length $y->{attr};
+                    next if length $x->{attr} == length $y->{attr} && $x->{attr} ge $y->{attr};
                 }
                 $drvPathToId{$x->{drvPath}} = $x;
             }
 
-            foreach my $job (values %{$jobs}) {
-                next unless $job->{constituents};
-
+            foreach my $job (values @jobsWithConstituents) {
+                next unless defined $job->{constituents};
                 if (defined $job->{error}) {
-                    die "aggregate job ‘$job->{jobName}’ failed with the error: $job->{error}\n";
+                    die "aggregate job ‘$job->{attr}’ failed with the error: $job->{error}\n";
                 }
 
                 my $x = $drvPathToId{$job->{drvPath}} or
-                    die "aggregate job ‘$job->{jobName}’ has no corresponding build record.\n";
+                    die "aggregate job ‘$job->{attr}’ has no corresponding build record.\n";
                 foreach my $drvPath (@{$job->{constituents}}) {
                     my $constituent = $drvPathToId{$drvPath};
                     if (defined $constituent) {
                         $db->resultset('AggregateConstituents')->update_or_create({aggregate => $x->{id}, constituent => $constituent->{id}});
                     } else {
-                        warn "aggregate job ‘$job->{jobName}’ has a constituent ‘$drvPath’ that doesn't correspond to a Hydra build\n";
+                        warn "aggregate job ‘$job->{attr}’ has a constituent ‘$drvPath’ that doesn't correspond to a Hydra build\n";
                     }
                 }
             }
@@ -857,11 +889,15 @@ sub checkJobsetWrapped {
         $jobset->update({ enabled => 0 }) if $jobset->enabled == 2;
 
         $jobset->update({ lastcheckedtime => time, forceeval => undef });
-    });
 
-    my $dbStop = clock_gettime(CLOCK_MONOTONIC);
+        $evaluationErrorRecord->update({ errormsg => $evaluationErrorMsg });
+        setJobsetError($jobset, $evaluationErrorMsg, $evaluationErrorTime);
 
-    Net::Statsd::timing("hydra.evaluator.db_time", int(($dbStop - $dbStart) * 1000));
+        $evalStop = clock_gettime(CLOCK_MONOTONIC);
+        $ev->update({ evaltime => abs(int($evalStop - $evalStart)) });
+    });
+
+    Net::Statsd::timing("hydra.evaluator.eval_time", int(($evalStop - $evalStart) * 1000));
     Net::Statsd::increment("hydra.evaluator.evals");
     Net::Statsd::increment("hydra.evaluator.cached_evals") unless $jobsetChanged;
 }
diff --git a/t/evaluator/evaluate-constituents-broken.t b/t/evaluator/evaluate-constituents-broken.t
index ed25d192d..0e5960bf6 100644
--- a/t/evaluator/evaluate-constituents-broken.t
+++ b/t/evaluator/evaluate-constituents-broken.t
@@ -18,14 +18,14 @@ isnt($res, 0, "hydra-eval-jobset exits non-zero");
 ok(utf8::decode($stderr), "Stderr output is UTF8-clean");
 like(
     $stderr,
-    qr/aggregate job ‘mixed_aggregate’ failed with the error: constituentA: does not exist/,
+    qr/aggregate job ‘mixed_aggregate’ failed with the error: "constituentA": does not exist/,
     "The stderr record includes a relevant error message"
 );
 
-$jobset->discard_changes;  # refresh from DB
+$jobset->discard_changes({ '+columns' => {'errormsg' => 'errormsg'} });  # refresh from DB
 like(
     $jobset->errormsg,
-    qr/aggregate job ‘mixed_aggregate’ failed with the error: constituentA: does not exist/,
+    qr/aggregate job ‘mixed_aggregate’ failed with the error: "constituentA": does not exist/,
     "The jobset records a relevant error message"
 );
 
diff --git a/t/evaluator/evaluate-meta.t b/t/evaluator/evaluate-meta.t
new file mode 100644
index 000000000..9f546a7f1
--- /dev/null
+++ b/t/evaluator/evaluate-meta.t
@@ -0,0 +1,22 @@
+use feature 'unicode_strings';
+use strict;
+use warnings;
+use Setup;
+use Test2::V0;
+
+my $ctx = test_context();
+
+my $builds = $ctx->makeAndEvaluateJobset(
+    expression => "meta.nix",
+    build => 1
+);
+
+my $build = $builds->{"full-of-meta"};
+
+is($build->finished, 1, "Build should be finished.");
+is($build->description, "This is the description of the job.", "Wrong description extracted from the build.");
+is($build->license, "MIT, BSD", "Wrong licenses extracted from the build.");
+is($build->homepage, "https://example.com/", "Wrong homepage extracted from the build.");
+is($build->maintainers, 'alice@example.com, bob@not.found', "Wrong maintainers extracted from the build.");
+
+done_testing;
diff --git a/t/jobs/meta.nix b/t/jobs/meta.nix
new file mode 100644
index 000000000..9204e3843
--- /dev/null
+++ b/t/jobs/meta.nix
@@ -0,0 +1,17 @@
+with import ./config.nix;
+{
+  full-of-meta =
+    mkDerivation {
+      name = "full-of-meta";
+      builder = ./empty-dir-builder.sh;
+
+      meta = {
+        description = "This is the description of the job.";
+        license = [ { shortName = "MIT"; } "BSD" ];
+        homepage = "https://example.com/";
+        maintainers = [ "alice@example.com" { email = "bob@not.found"; } ];
+
+        outPath = "${placeholder "out"}";
+      };
+    };
+}
diff --git a/t/meson.build b/t/meson.build
index 11044a03e..c3c58458d 100644
--- a/t/meson.build
+++ b/t/meson.build
@@ -27,7 +27,6 @@ testenv.prepend('PERL5LIB',
   separator: ':'
 )
 testenv.prepend('PATH',
-  fs.parent(hydra_eval_jobs.full_path()),
   fs.parent(hydra_evaluator.full_path()),
   fs.parent(hydra_queue_runner.full_path()),
   meson.project_source_root() / 'src/script',
diff --git a/t/queue-runner/constituents.t b/t/queue-runner/constituents.t
index c63336420..e1b8d733f 100644
--- a/t/queue-runner/constituents.t
+++ b/t/queue-runner/constituents.t
@@ -22,11 +22,11 @@ is(nrQueuedBuildsForJobset($jobset), 0, "Evaluating jobs/broken-constituent.nix
 
 like(
     $jobset->errormsg,
-    qr/^does-not-exist: does not exist$/m,
+    qr/^"does-not-exist": does not exist$/m,
     "Evaluating jobs/broken-constituent.nix should log an error for does-not-exist");
 like(
     $jobset->errormsg,
-    qr/^does-not-evaluate: error: assertion 'false' failed$/m,
+    qr/^"does-not-evaluate": "error: assertion 'false' failed/m,
     "Evaluating jobs/broken-constituent.nix should log an error for does-not-evaluate");
 
 done_testing;